Exercise 1 Solutions

This exercise will involve estimating causal effect parameters using a difference-in-differences identification strategy that involves conditioning on covariates in the parallel trends assumption and possibly allows for anticipation effects.

In particular, we will use data from the National Longitudinal Study of Youth to learn about causal effects of job displacement (where job displacement roughly means “losing your job through no fault of your own” — a mass layoff is a main example).

To start with, load the data from the file job_displacement_data.RData by running

use "../job_displacement_data.dta", clear

which will load a dataset called job_displacement_data. This is what the data looks like

list in 1/5
     |      id   year   group   income   female   white   occ_sc~e |
     |-------------------------------------------------------------|
  1. | 7900002   1984       0    31130        1       1          4 |
  2. | 7900002   1985       0    32200        1       1          3 |
  3. | 7900002   1986       0    35520        1       1          4 |
  4. | 7900002   1987       0    43600        1       1          4 |
  5. | 7900002   1988       0    39900        1       1          4 |
     +-------------------------------------------------------------+

You can see that the data contains the following columns:

  • id - an individual identifier
  • year - the year for this observation
  • group - the year that person lost his/her job. group=0 for those that do not lose a job in any period being considered.
  • income - a person’s wage and salary income in this year
  • female - 1 for females, 0 for males
  • white - 1 for white, 0 for non-white

For the results below, we will mainly use the csdid package which you can install using ssc install csdid.

Question 1

We will start by computing group-time average treatment effects without including any covariates in the parallel trends assumption.

  1. Use the did package to compute all available group-time average treatment effects.
Solutions:
csdid income, ivar(id) time(year) gvar(group)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,400
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
g1985        |
 t_1984_1985 |  -9455.758   3530.143    -2.68   0.007    -16374.71   -2536.805
 t_1984_1986 |  -14981.15   4299.966    -3.48   0.000    -23408.93   -6553.376
 t_1984_1987 |  -6129.213   4337.391    -1.41   0.158    -14630.34    2371.917
 t_1984_1988 |  -4815.918   4738.082    -1.02   0.309    -14102.39    4470.551
 t_1984_1989 |          0  (omitted)
 t_1984_1990 |  -8011.917   5687.048    -1.41   0.159    -19158.33    3134.491
 t_1984_1991 |  -8164.492   5878.675    -1.39   0.165    -19686.48    3357.498
 t_1984_1992 |  -6325.888   5590.747    -1.13   0.258    -17283.55    4631.775
 t_1984_1993 |  -9669.584   5724.552    -1.69   0.091     -20889.5    1550.332
-------------+----------------------------------------------------------------
g1986        |
 t_1984_1985 |  -1801.937    2456.11    -0.73   0.463    -6615.824     3011.95
 t_1985_1986 |  -1919.447   3405.188    -0.56   0.573    -8593.493    4754.598
 t_1985_1987 |  -2596.819   4304.758    -0.60   0.546    -11033.99    5840.353
 t_1985_1988 |  -2081.753   6447.175    -0.32   0.747    -14717.98    10554.48
 t_1985_1989 |          0  (omitted)
 t_1985_1990 |  -6064.094   6179.644    -0.98   0.326    -18175.97    6047.785
 t_1985_1991 |  -5903.964    6329.81    -0.93   0.351    -18310.16    6502.237
 t_1985_1992 |  -6804.483    6558.35    -1.04   0.299    -19658.61    6049.647
 t_1985_1993 |  -1801.576   6383.008    -0.28   0.778    -14312.04    10708.89
-------------+----------------------------------------------------------------
g1987        |
 t_1984_1985 |   4518.574    4564.82     0.99   0.322    -4428.308    13465.46
 t_1985_1986 |  -8012.488   4349.707    -1.84   0.065    -16537.76    512.7802
 t_1986_1987 |   7048.857   6144.013     1.15   0.251    -4993.188     19090.9
 t_1986_1988 |   4489.467   6171.365     0.73   0.467    -7606.187    16585.12
 t_1986_1989 |          0  (omitted)
 t_1986_1990 |   8004.136   6887.031     1.16   0.245    -5494.197    21502.47
 t_1986_1991 |   9475.066   6911.544     1.37   0.170    -4071.312    23021.44
 t_1986_1992 |   8533.541   9383.704     0.91   0.363    -9858.181    26925.26
 t_1986_1993 |   7881.393   7250.427     1.09   0.277    -6329.182    22091.97
-------------+----------------------------------------------------------------
g1988        |
 t_1984_1985 |  -8350.771   4329.706    -1.93   0.054    -16836.84    135.2963
 t_1985_1986 |  -3420.853   2964.689    -1.15   0.249    -9231.537    2389.831
 t_1986_1987 |  -3617.674   3483.742    -1.04   0.299    -10445.68    3210.334
 t_1987_1988 |  -1173.817   2850.037    -0.41   0.680    -6759.787    4412.153
 t_1987_1989 |          0  (omitted)
 t_1987_1990 |   280.6263    5519.59     0.05   0.959    -10537.57    11098.82
 t_1987_1991 |   6099.727   4026.311     1.51   0.130    -1791.697    13991.15
 t_1987_1992 |   13737.82   10419.28     1.32   0.187    -6683.587    34159.22
 t_1987_1993 |   1688.782    7747.27     0.22   0.827    -13495.59    16873.15
-------------+----------------------------------------------------------------
g1990        |
 t_1984_1985 |  -5281.536   3137.971    -1.68   0.092    -11431.85    868.7732
 t_1985_1986 |   3654.173   2446.867     1.49   0.135    -1141.598    8449.943
 t_1986_1987 |   5934.895   2948.335     2.01   0.044     156.2642    11713.53
 t_1987_1988 |   1034.199   3133.832     0.33   0.741        -5108    7176.398
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1989_1991 |          0  (omitted)
 t_1989_1992 |          0  (omitted)
 t_1989_1993 |          0  (omitted)
-------------+----------------------------------------------------------------
g1991        |
 t_1984_1985 |   891.2874   2765.972     0.32   0.747    -4529.918    6312.492
 t_1985_1986 |  -2816.636   3299.083    -0.85   0.393    -9282.719    3649.448
 t_1986_1987 |  -1340.055   2532.177    -0.53   0.597     -6303.03     3622.92
 t_1987_1988 |  -7025.039   3544.277    -1.98   0.047    -13971.69   -78.38372
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |  -12150.64   3997.579    -3.04   0.002    -19985.76   -4315.534
 t_1990_1992 |   1433.998   4139.233     0.35   0.729    -6678.749    9546.745
 t_1990_1993 |  -2679.828   6842.388    -0.39   0.695    -16090.66    10731.01
-------------+----------------------------------------------------------------
g1992        |
 t_1984_1985 |  -12110.06   6253.041    -1.94   0.053    -24365.79    145.6789
 t_1985_1986 |  -3287.561   2324.793    -1.41   0.157    -7844.072    1268.951
 t_1986_1987 |   2300.028   3450.526     0.67   0.505    -4462.878    9062.935
 t_1987_1988 |  -7273.935   2434.951    -2.99   0.003    -12046.35   -2501.517
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |   -10031.7   7303.289    -1.37   0.170    -24345.89    4282.481
 t_1991_1992 |   -8990.85    3612.76    -2.49   0.013    -16071.73   -1909.971
 t_1991_1993 |  -8662.612   12070.67    -0.72   0.473    -32320.69    14995.47
-------------+----------------------------------------------------------------
g1993        |
 t_1984_1985 |  -7424.664   4439.089    -1.67   0.094    -16125.12     1275.79
 t_1985_1986 |    677.906   2503.711     0.27   0.787    -4229.277    5585.089
 t_1986_1987 |   1424.138   2921.033     0.49   0.626    -4300.981    7149.258
 t_1987_1988 |   4778.256   1527.433     3.13   0.002     1784.542    7771.969
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |   3664.883   4980.086     0.74   0.462    -6095.907    13425.67
 t_1991_1992 |  -4108.917   4427.656    -0.93   0.353    -12786.96     4569.13
 t_1992_1993 |  -22828.36   5126.199    -4.45   0.000    -32875.53    -12781.2
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details
  1. Bonus Question Try to manually calculate \(ATT(g=1992, t=1992)\). Can you calculate exactly the same number as in part (a)?
Solutions:
quietly sum income if group == 1992 & year == 1992
local y_11 = `r(mean)'
quietly sum income if group == 1992 & year == 1991
local y_10 = `r(mean)'
quietly sum income if group == 0 & year == 1992
local y_01 = `r(mean)'
quietly sum income if group == 0 & year == 1991
local y_00 = `r(mean)'
local did = (`y_11' - `y_10') - (`y_01' - `y_00')
disp "DID Estimate: `did'"
DID Estimate: -8990.850446338249
  1. Aggregate the group-time average treatment effects into an event study and plot the results. What do you notice? Is there evidence against parallel trends?
Solutions:
csdid income, ivar(id) time(year) gvar(group) agg(event)
csdid_plot
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,400
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Pre_avg |  -2730.008   1048.027    -2.60   0.009    -4784.103   -675.9126
    Post_avg |  -3595.643   2811.038    -1.28   0.201    -9105.176     1913.89
         Tm8 |  -7424.664   4439.089    -1.67   0.094    -16125.12     1275.79
         Tm7 |  -5716.076   3838.157    -1.49   0.136    -13238.72    1806.573
         Tm6 |  -202.5117    1629.01    -0.12   0.901    -3395.313    2990.289
         Tm5 |  -1357.557   1743.251    -0.78   0.436    -4774.267    2059.153
         Tm4 |  -404.3836   1695.983    -0.24   0.812    -3728.449    2919.682
         Tm3 |  -2834.594   2475.856    -1.14   0.252    -7687.181    2017.994
         Tm2 |   943.6407   1953.637     0.48   0.629    -2885.417    4772.698
         Tm1 |   -4843.92   1829.255    -2.65   0.008    -8429.194   -1258.646
         Tp0 |  -5970.064   1834.463    -3.25   0.001    -9565.546   -2374.582
         Tp1 |  -6610.944   2700.649    -2.45   0.014    -11904.12   -1317.769
         Tp2 |  -3447.418   2827.528    -1.22   0.223    -8989.271    2094.435
         Tp3 |   624.9279   3225.022     0.19   0.846    -5695.999    6945.855
         Tp4 |   4919.383   4866.129     1.01   0.312    -4618.055    14456.82
         Tp5 |  -2941.344     3614.3    -0.81   0.416    -10025.24    4142.554
         Tp6 |  -4518.106   3968.625    -1.14   0.255    -12296.47    3260.255
         Tp7 |  -4747.639   4302.736    -1.10   0.270    -13180.85    3685.568
         Tp8 |  -9669.584   5724.552    -1.69   0.091     -20889.5    1550.332
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

  1. Aggregate the group-time average treatment effects into a single overall treatment effect. How do you interpret the results?
Solutions:
csdid income, ivar(id) time(year) gvar(group) agg(group)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,400
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    GAverage |  -4406.669   2143.766    -2.06   0.040    -8608.374   -204.9649
       G1985 |  -8444.241   4535.853    -1.86   0.063    -17334.35    445.8679
       G1986 |  -3881.734   5257.768    -0.74   0.460    -14186.77    6423.302
       G1987 |   7572.077   6130.068     1.24   0.217    -4442.637    19586.79
       G1988 |   4126.627   4583.124     0.90   0.368    -4856.131    13109.39
       G1991 |  -4465.492   4514.597    -0.99   0.323    -13313.94    4382.956
       G1992 |  -8826.731   6562.055    -1.35   0.179    -21688.12    4034.661
       G1993 |  -22828.36   5126.199    -4.45   0.000    -32875.53    -12781.2
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

Question 2

A major issue in the job displacement literature concerns a version of anticipation. In particular, there is some empirical evidence that earnings of displaced workers start to decline before they are actually displaced (a rough explanation is that firms where there are mass layoffs typically “struggle” in the time period before the mass layoff actually takes place and this can lead to slower income growth for workers at those firms).

  1. Is there evidence of anticipation in your results from Question 1?
Solutions:

There is a moderate amount of evidence for anticipation in the previous results. It hinges on the estimate for event-time equal to -1. It is negative which is in line with the discussion about anticipation above, but it is only marginally statistically significant.

  1. Repeat parts (a)-(d) of Question 1 allowing for one year of anticipation.
Solutions:
* Move up "treatment date" by 1 year
gen group_m1 = group
replace group_m1 = group_m1 - 1 if group != 0
(1,125 real changes made)
* part a
csdid income, ivar(id) time(year) gvar(group_m1)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
g1985        |
 t_1984_1985 |  -1801.937    2456.11    -0.73   0.463    -6615.824     3011.95
 t_1984_1986 |  -3721.385   3344.837    -1.11   0.266    -10277.15    2834.376
 t_1984_1987 |  -4398.756   3453.298    -1.27   0.203     -11167.1    2369.583
 t_1984_1988 |  -3883.691   5544.658    -0.70   0.484    -14751.02     6983.64
 t_1984_1989 |          0  (omitted)
 t_1984_1990 |  -7866.031   5527.045    -1.42   0.155    -18698.84    2966.778
 t_1984_1991 |  -7705.901   5713.436    -1.35   0.177    -18904.03    3492.228
 t_1984_1992 |  -8606.421   6086.643    -1.41   0.157    -20536.02    3323.181
 t_1984_1993 |  -3603.513   5529.124    -0.65   0.515     -14440.4    7233.371
-------------+----------------------------------------------------------------
g1986        |
 t_1984_1985 |   4518.574    4564.82     0.99   0.322    -4428.308    13465.46
 t_1985_1986 |  -8012.488   4349.707    -1.84   0.065    -16537.76    512.7802
 t_1985_1987 |  -963.6314   6479.146    -0.15   0.882    -13662.52    11735.26
 t_1985_1988 |  -3523.021   7396.681    -0.48   0.634    -18020.25    10974.21
 t_1985_1989 |          0  (omitted)
 t_1985_1990 |  -8.351815    6222.58    -0.00   0.999    -12204.38    12187.68
 t_1985_1991 |   1462.578   6809.271     0.21   0.830    -11883.35     14808.5
 t_1985_1992 |   521.0534   8752.718     0.06   0.953    -16633.96    17676.07
 t_1985_1993 |  -131.0948    6942.23    -0.02   0.985    -13737.62    13475.43
-------------+----------------------------------------------------------------
g1987        |
 t_1984_1985 |  -8350.771   4329.706    -1.93   0.054    -16836.84    135.2963
 t_1985_1986 |  -3420.853   2964.689    -1.15   0.249    -9231.537    2389.831
 t_1986_1987 |  -3617.674   3483.742    -1.04   0.299    -10445.68    3210.334
 t_1986_1988 |  -4791.491    4153.77    -1.15   0.249    -12932.73    3349.749
 t_1986_1989 |          0  (omitted)
 t_1986_1990 |  -3337.048   6051.583    -0.55   0.581    -15197.93    8523.837
 t_1986_1991 |   2482.053   5357.229     0.46   0.643    -8017.922    12982.03
 t_1986_1992 |   10120.14   11738.47     0.86   0.389    -12886.84    33127.13
 t_1986_1993 |  -1928.892   7250.619    -0.27   0.790    -16139.85    12282.06
-------------+----------------------------------------------------------------
g1989        |
 t_1984_1985 |  -5281.536   3137.971    -1.68   0.092    -11431.85    868.7732
 t_1985_1986 |   3654.173   2446.867     1.49   0.135    -1141.598    8449.943
 t_1986_1987 |   5934.895   2948.335     2.01   0.044     156.2642    11713.53
 t_1987_1988 |   1034.199   3133.832     0.33   0.741        -5108    7176.398
 t_1988_1989 |          0  (omitted)
 t_1988_1990 |  -4343.949   9169.925    -0.47   0.636    -22316.67    13628.77
 t_1988_1991 |  -21910.21   4407.569    -4.97   0.000    -30548.89   -13271.53
 t_1988_1992 |  -15365.93   3710.792    -4.14   0.000    -22638.95   -8092.909
 t_1988_1993 |  -16411.11   6044.992    -2.71   0.007    -28259.07   -4563.139
-------------+----------------------------------------------------------------
g1990        |
 t_1984_1985 |   891.2874   2765.972     0.32   0.747    -4529.918    6312.492
 t_1985_1986 |  -2816.636   3299.083    -0.85   0.393    -9282.719    3649.448
 t_1986_1987 |  -1340.055   2532.177    -0.53   0.597     -6303.03     3622.92
 t_1987_1988 |  -7025.039   3544.277    -1.98   0.047    -13971.69   -78.38372
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1989_1991 |          0  (omitted)
 t_1989_1992 |          0  (omitted)
 t_1989_1993 |          0  (omitted)
-------------+----------------------------------------------------------------
g1991        |
 t_1984_1985 |  -12110.06   6253.041    -1.94   0.053    -24365.79    145.6789
 t_1985_1986 |  -3287.561   2324.793    -1.41   0.157    -7844.072    1268.951
 t_1986_1987 |   2300.028   3450.526     0.67   0.505    -4462.878    9062.935
 t_1987_1988 |  -7273.935   2434.951    -2.99   0.003    -12046.35   -2501.517
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |   -10031.7   7303.289    -1.37   0.170    -24345.89    4282.481
 t_1990_1992 |  -19022.55   6414.226    -2.97   0.003     -31594.2   -6450.902
 t_1990_1993 |  -18694.31   7778.507    -2.40   0.016    -33939.91   -3448.721
-------------+----------------------------------------------------------------
g1992        |
 t_1984_1985 |  -7424.664   4439.089    -1.67   0.094    -16125.12     1275.79
 t_1985_1986 |    677.906   2503.711     0.27   0.787    -4229.277    5585.089
 t_1986_1987 |   1424.138   2921.033     0.49   0.626    -4300.981    7149.258
 t_1987_1988 |   4778.256   1527.433     3.13   0.002     1784.542    7771.969
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |   3664.883   4980.086     0.74   0.462    -6095.907    13425.67
 t_1991_1992 |  -4108.917   4427.656    -0.93   0.353    -12786.96     4569.13
 t_1991_1993 |  -26937.28   5505.881    -4.89   0.000    -37728.61   -16145.95
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details
* part b
quietly sum income if group == 1992 & year == 1992
local y_11 = `r(mean)'
quietly sum income if group == 1992 & year == 1990
local y_10 = `r(mean)'
quietly sum income if group == 0 & year == 1992
local y_01 = `r(mean)'
quietly sum income if group == 0 & year == 1990
local y_00 = `r(mean)'
local did = (`y_11' - `y_10') - (`y_01' - `y_00')
disp "DID Estimate: `did'"
DID Estimate: -19022.55321626245
* part c
csdid income, ivar(id) time(year) gvar(group_m1) agg(event)
csdid_plot
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Pre_avg |  -2428.021   1129.103    -2.15   0.032    -4641.022   -215.0193
    Post_avg |  -5116.378   2522.756    -2.03   0.043    -10060.89   -171.8674
         Tm7 |  -7424.664   4439.089    -1.67   0.094    -16125.12     1275.79
         Tm6 |  -5716.076   3838.157    -1.49   0.136    -13238.72    1806.573
         Tm5 |  -202.5117    1629.01    -0.12   0.901    -3395.313    2990.289
         Tm4 |  -1357.557   1743.251    -0.78   0.436    -4774.267    2059.153
         Tm3 |  -404.3836   1695.983    -0.24   0.812    -3728.449    2919.682
         Tm2 |  -2834.594   2475.856    -1.14   0.252    -7687.181    2017.994
         Tm1 |   943.6407   1953.637     0.48   0.629    -2885.417    4772.698
         Tp0 |   -4843.92   1829.255    -2.65   0.008    -8429.194   -1258.646
         Tp1 |  -7247.563   2707.316    -2.68   0.007    -12553.81    -1941.32
         Tp2 |  -10905.07   2995.137    -3.64   0.000    -16775.43   -5034.707
         Tp3 |  -7150.703   3193.771    -2.24   0.025    -13410.38   -891.0268
         Tp4 |  -4576.623   3674.477    -1.25   0.213    -11778.47    2625.219
         Tp5 |    760.506   5021.692     0.15   0.880     -9081.83    10602.84
         Tp6 |  -3459.808    4160.34    -0.83   0.406    -11613.92    4694.307
         Tp7 |  -5020.706   4673.388    -1.07   0.283    -14180.38    4138.967
         Tp8 |  -3603.513   5529.124    -0.65   0.515     -14440.4    7233.371
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

* part d
csdid income, ivar(id) time(year) gvar(group_m1) agg(group)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
    GAverage |  -7298.004   1967.793    -3.71   0.000    -11154.81   -3441.201
       G1985 |  -5198.454   3973.626    -1.31   0.191    -12986.62    2589.709
       G1986 |  -1522.137   5439.282    -0.28   0.780    -12182.93     9138.66
       G1987 |  -178.8183   4986.896    -0.04   0.971    -9952.954    9595.317
       G1989 |   -14507.8   4121.479    -3.52   0.000    -22585.75   -6429.847
       G1991 |  -15916.19    4228.73    -3.76   0.000    -24204.35   -7628.033
       G1992 |   -15523.1   4288.351    -3.62   0.000    -23928.11   -7118.085
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

Question 3

Now, let’s suppose that we think that parallel trends holds only after we condition on a person sex and race (in reality, you could think of including many other variables in the parallel trends assumption, but let’s just keep it simple). In my view, I think allowing for anticipation is desirable in this setting too, so let’s keep allowing for one year of anticipation.

  1. Answer parts (a), (c), and (d) of Question 1 but including sex and white as covariates.
Solutions:
* part a
csdid income i.female i.white, ivar(id) time(year) gvar(group_m1)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : least squares
Treatment model: inverse probability
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
g1985        |
 t_1984_1985 |  -1724.003   2458.934    -0.70   0.483    -6543.425    3095.418
 t_1984_1986 |  -4258.867   3329.635    -1.28   0.201    -10784.83    2267.097
 t_1984_1987 |  -4861.614   3475.476    -1.40   0.162    -11673.42    1950.195
 t_1984_1988 |  -4729.612   5484.551    -0.86   0.388    -15479.14    6019.911
 t_1984_1989 |          0  (omitted)
 t_1984_1990 |   -8685.99   5550.638    -1.56   0.118    -19565.04     2193.06
 t_1984_1991 |  -8753.855   5698.922    -1.54   0.125    -19923.54    2415.826
 t_1984_1992 |  -9530.395    6046.17    -1.58   0.115    -21380.67    2319.881
 t_1984_1993 |  -4727.765   5426.184    -0.87   0.384    -15362.89    5907.361
-------------+----------------------------------------------------------------
g1986        |
 t_1984_1985 |   4559.705   4596.627     0.99   0.321    -4449.519    13568.93
 t_1985_1986 |   -8337.68   4317.782    -1.93   0.053    -16800.38    125.0165
 t_1985_1987 |  -1244.485   6506.007    -0.19   0.848    -13996.02    11507.05
 t_1985_1988 |  -4009.114   7416.923    -0.54   0.589    -18546.02    10527.79
 t_1985_1989 |          0  (omitted)
 t_1985_1990 |  -483.2506   6321.764    -0.08   0.939    -12873.68    11907.18
 t_1985_1991 |   865.8558   6863.468     0.13   0.900    -12586.29    14318.01
 t_1985_1992 |  -1.136882   8750.712    -0.00   1.000    -17152.22    17149.94
 t_1985_1993 |  -760.5834   6996.293    -0.11   0.913    -14473.06     12951.9
-------------+----------------------------------------------------------------
g1987        |
 t_1984_1985 |  -8427.959   4344.236    -1.94   0.052    -16942.51    86.58735
 t_1985_1986 |  -3208.663   2967.399    -1.08   0.280    -9024.659    2607.332
 t_1986_1987 |  -3540.335   3532.861    -1.00   0.316    -10464.61    3383.945
 t_1986_1988 |  -4496.718   4192.765    -1.07   0.283    -12714.39    3720.951
 t_1986_1989 |          0  (omitted)
 t_1986_1990 |  -2886.271   6080.348    -0.47   0.635    -14803.53    9030.992
 t_1986_1991 |   3026.129   5380.889     0.56   0.574    -7520.219    13572.48
 t_1986_1992 |   10422.75   11781.13     0.88   0.376    -12667.85    33513.35
 t_1986_1993 |  -1710.323   7296.294    -0.23   0.815     -16010.8    12590.15
-------------+----------------------------------------------------------------
g1989        |
 t_1984_1985 |  -5423.422   3167.925    -1.71   0.087    -11632.44     785.597
 t_1985_1986 |   4124.357   2580.371     1.60   0.110     -933.078    9181.792
 t_1986_1987 |    6034.51   2986.352     2.02   0.043     181.3663    11887.65
 t_1987_1988 |   1473.845   3224.853     0.46   0.648    -4846.751    7794.441
 t_1988_1989 |          0  (omitted)
 t_1988_1990 |   -4087.09     9194.5    -0.44   0.657    -22107.98     13933.8
 t_1988_1991 |  -21451.71   4426.529    -4.85   0.000    -30127.55   -12775.87
 t_1988_1992 |  -15350.47   3712.715    -4.13   0.000    -22627.26   -8073.681
 t_1988_1993 |  -16489.87   6082.825    -2.71   0.007    -28411.98   -4567.748
-------------+----------------------------------------------------------------
g1990        |
 t_1984_1985 |   787.4357    2762.92     0.29   0.776    -4627.788    6202.659
 t_1985_1986 |  -2463.712   3387.522    -0.73   0.467    -9103.133    4175.708
 t_1986_1987 |  -1271.944   2604.541    -0.49   0.625    -6376.751    3832.863
 t_1987_1988 |  -6698.783   3730.878    -1.80   0.073    -14011.17    613.6032
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1989_1991 |          0  (omitted)
 t_1989_1992 |          0  (omitted)
 t_1989_1993 |          0  (omitted)
-------------+----------------------------------------------------------------
g1991        |
 t_1984_1985 |  -12170.12   6178.935    -1.97   0.049    -24280.61   -59.62972
 t_1985_1986 |  -3584.494   2342.734    -1.53   0.126    -8176.168     1007.18
 t_1986_1987 |   2598.525   3439.103     0.76   0.450    -4141.994    9339.043
 t_1987_1988 |  -7330.915   2607.234    -2.81   0.005       -12441    -2220.83
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |  -10130.91   7396.362    -1.37   0.171    -24627.52    4365.689
 t_1990_1992 |   -19327.8   6455.656    -2.99   0.003    -31980.65   -6674.944
 t_1990_1993 |  -19410.44   7667.027    -2.53   0.011    -34437.54   -4383.345
-------------+----------------------------------------------------------------
g1992        |
 t_1984_1985 |  -7391.929   4506.159    -1.64   0.101    -16223.84    1439.981
 t_1985_1986 |   50.76363    2681.48     0.02   0.985    -5204.841    5306.368
 t_1986_1987 |   1618.304   2902.035     0.56   0.577    -4069.581    7306.189
 t_1987_1988 |   4453.454   1642.082     2.71   0.007     1235.032    7671.877
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |   3439.787   5033.438     0.68   0.494     -6425.57    13305.14
 t_1991_1992 |  -4123.758   4592.035    -0.90   0.369    -13123.98    4876.466
 t_1991_1993 |  -27304.41   5673.621    -4.81   0.000     -38424.5   -16184.32
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details
* part c
csdid income i.female i.white, ivar(id) time(year) gvar(group_m1) agg(event)
csdid_plot
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : least squares
Treatment model: inverse probability
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Pre_avg |  -2409.952   1129.233    -2.13   0.033    -4623.208   -196.6955
    Post_avg |  -5488.493   2523.002    -2.18   0.030    -10433.49   -543.5007
         Tm7 |  -7391.929   4506.159    -1.64   0.101    -16223.84    1439.981
         Tm6 |  -6059.679    3798.62    -1.60   0.111    -13504.84     1385.48
         Tm5 |  -274.8826    1629.93    -0.17   0.866    -3469.487    2919.722
         Tm4 |  -1327.465   1760.053    -0.75   0.451    -4777.105    2122.174
         Tm3 |  -179.5675   1777.944    -0.10   0.920    -3664.273    3305.138
         Tm2 |  -2749.685   2504.009    -1.10   0.272    -7657.453    2158.082
         Tm1 |   1113.547   1970.672     0.57   0.572    -2748.899    4975.994
         Tp0 |  -4884.842   1846.201    -2.65   0.008    -8503.329   -1266.354
         Tp1 |  -7378.857   2720.712    -2.71   0.007    -12711.35    -2046.36
         Tp2 |  -11156.99   2979.612    -3.74   0.000    -16996.92   -5317.058
         Tp3 |  -7316.783   3181.754    -2.30   0.021    -13552.91    -1080.66
         Tp4 |  -4551.513   3709.318    -1.23   0.220    -11821.64    2718.616
         Tp5 |   377.7002   5063.744     0.07   0.941    -9547.056    10302.46
         Tp6 |  -3937.296   4177.605    -0.94   0.346    -12125.25    4250.659
         Tp7 |   -5820.09   4683.423    -1.24   0.214    -14999.43     3359.25
         Tp8 |  -4727.765   5426.184    -0.87   0.384    -15362.89    5907.361
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

* part d
csdid income i.female i.white, ivar(id) time(year) gvar(group_,1) agg(group)
gvar() does not contain a valid varname
r(198);

end of do-file
r(198);
  1. By default, the did package uses the doubly robust approach that we discussed during our session. How do the results change if you use a regression approach or propensity score re-weighting?
Solutions:

For simplicity, I am just going to show the overall results when using the regression approach and the propensity score re-weighting approach.

* part a
csdid income i.female i.white, ivar(id) time(year) gvar(group_m1) method(reg)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
g1985        |
 t_1984_1985 |  -1731.595    2458.97    -0.70   0.481    -6551.087    3087.897
 t_1984_1986 |  -4223.039   3336.126    -1.27   0.206    -10761.73    2315.649
 t_1984_1987 |  -4807.033   3469.401    -1.39   0.166    -11606.94    1992.868
 t_1984_1988 |  -4656.409   5490.373    -0.85   0.396    -15417.34    6104.525
 t_1984_1989 |          0  (omitted)
 t_1984_1990 |  -8618.812   5546.257    -1.55   0.120    -19489.27    2251.652
 t_1984_1991 |  -8675.577   5697.411    -1.52   0.128     -19842.3    2491.142
 t_1984_1992 |  -9431.255   6055.712    -1.56   0.119    -21300.23    2437.724
 t_1984_1993 |  -4626.267   5442.481    -0.85   0.395    -15293.33      6040.8
-------------+----------------------------------------------------------------
g1986        |
 t_1984_1985 |   4557.422   4594.569     0.99   0.321    -4447.768    13562.61
 t_1985_1986 |  -8324.622   4315.666    -1.93   0.054    -16783.17    133.9279
 t_1985_1987 |  -1225.788    6500.99    -0.19   0.850    -13967.49    11515.92
 t_1985_1988 |  -3984.817    7409.26    -0.54   0.591     -18506.7    10537.07
 t_1985_1989 |          0  (omitted)
 t_1985_1990 |  -460.7647   6311.823    -0.07   0.942    -12831.71    11910.18
 t_1985_1991 |   891.6797   6855.723     0.13   0.897    -12545.29    14328.65
 t_1985_1992 |   30.96108    8742.63     0.00   0.997    -17104.28     17166.2
 t_1985_1993 |  -727.7764    6980.71    -0.10   0.917    -14409.72    12954.16
-------------+----------------------------------------------------------------
g1987        |
 t_1984_1985 |  -8426.549    4341.95    -1.94   0.052    -16936.61    83.51623
 t_1985_1986 |  -3216.731   2983.078    -1.08   0.281    -9063.457    2629.995
 t_1986_1987 |  -3543.819   3523.884    -1.01   0.315     -10450.5    3362.867
 t_1986_1988 |  -4503.662   4176.624    -1.08   0.281    -12689.69     3682.37
 t_1986_1989 |          0  (omitted)
 t_1986_1990 |  -2892.095   6072.574    -0.48   0.634    -14794.12    9009.931
 t_1986_1991 |   3018.242   5364.143     0.56   0.574    -7495.286    13531.77
 t_1986_1992 |   10410.99   11755.85     0.89   0.376    -12630.06    33452.03
 t_1986_1993 |  -1722.525   7275.433    -0.24   0.813    -15982.11    12537.06
-------------+----------------------------------------------------------------
g1989        |
 t_1984_1985 |  -5423.935   3175.234    -1.71   0.088    -11647.28    799.4096
 t_1985_1986 |   4127.288   2600.393     1.59   0.112    -969.3893    9223.965
 t_1986_1987 |   6035.775   2972.772     2.03   0.042     209.2485     11862.3
 t_1987_1988 |   1475.102   3240.398     0.46   0.649    -4875.961    7826.165
 t_1988_1989 |          0  (omitted)
 t_1988_1990 |  -4087.497   9195.523    -0.44   0.657    -22110.39     13935.4
 t_1988_1991 |  -21451.37   4427.528    -4.84   0.000    -30129.16   -12773.57
 t_1988_1992 |  -15348.72   3740.808    -4.10   0.000    -22680.57   -8016.868
 t_1988_1993 |  -16487.96   6101.742    -2.70   0.007    -28447.15    -4528.76
-------------+----------------------------------------------------------------
g1990        |
 t_1984_1985 |   786.6699    2758.66     0.29   0.776    -4620.205    6193.545
 t_1985_1986 |  -2459.333   3400.778    -0.72   0.470    -9124.735     4206.07
 t_1986_1987 |  -1270.053   2587.731    -0.49   0.624    -6341.912    3801.807
 t_1987_1988 |  -6696.904   3730.782    -1.80   0.073     -14009.1    615.2933
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1989_1991 |          0  (omitted)
 t_1989_1992 |          0  (omitted)
 t_1989_1993 |          0  (omitted)
-------------+----------------------------------------------------------------
g1991        |
 t_1984_1985 |  -12153.64   6196.088    -1.96   0.050    -24297.75   -9.529399
 t_1985_1986 |  -3678.766   2417.222    -1.52   0.128    -8416.435    1058.902
 t_1986_1987 |   2557.812   3438.055     0.74   0.457    -4180.652    9296.276
 t_1987_1988 |  -7371.348   2582.541    -2.85   0.004    -12433.04    -2309.66
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |  -10155.01   7386.076    -1.37   0.169    -24631.46     4321.43
 t_1990_1992 |  -19397.19    6456.32    -3.00   0.003    -32051.35   -6743.037
 t_1990_1993 |  -19484.96   7660.212    -2.54   0.011    -34498.69   -4471.217
-------------+----------------------------------------------------------------
g1992        |
 t_1984_1985 |  -7392.683   4496.647    -1.64   0.100    -16205.95    1420.583
 t_1985_1986 |   55.07588   2720.682     0.02   0.984    -5277.363    5387.515
 t_1986_1987 |   1620.166   2926.908     0.55   0.580    -4116.469    7356.802
 t_1987_1988 |   4455.304   1639.589     2.72   0.007     1241.769    7668.839
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |    3440.89   5020.864     0.69   0.493    -6399.823     13281.6
 t_1991_1992 |  -4121.686   4628.339    -0.89   0.373    -13193.06    4949.692
 t_1991_1993 |   -27302.1   5691.967    -4.80   0.000    -38458.15   -16146.05
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details
* part c
csdid income i.female i.white, ivar(id) time(year) gvar(group_m1) agg(event) method(reg)
csdid_plot
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : regression adjustment
Treatment model: none
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Pre_avg |  -2414.705   1130.956    -2.14   0.033    -4631.338   -198.0717
    Post_avg |  -5457.024   2521.645    -2.16   0.030    -10399.36   -514.6912
         Tm7 |  -7392.683   4496.647    -1.64   0.100    -16205.95    1420.583
         Tm6 |  -6049.281   3811.814    -1.59   0.113     -13520.3    1421.736
         Tm5 |   -302.912   1645.723    -0.18   0.854    -3528.469    2922.646
         Tm4 |  -1333.849   1763.177    -0.76   0.449    -4789.612    2121.913
         Tm3 |  -186.9635   1781.588    -0.10   0.916    -3678.811    3304.884
         Tm2 |  -2748.214    2501.22    -1.10   0.272    -7650.516    2154.088
         Tm1 |   1110.969   1974.156     0.56   0.574    -2758.306    4980.244
         Tp0 |  -4887.737   1845.509    -2.65   0.008    -8504.869   -1270.606
         Tp1 |  -7374.962    2720.81    -2.71   0.007    -12707.65   -2042.273
         Tp2 |  -11142.38   2979.227    -3.74   0.000    -16981.56   -5303.199
         Tp3 |    -7290.7   3185.067    -2.29   0.022    -13533.32   -1048.083
         Tp4 |  -4546.854   3708.065    -1.23   0.220    -11814.53    2720.821
         Tp5 |   406.9005   5054.224     0.08   0.936    -9499.196       10313
         Tp6 |  -3902.203   4171.008    -0.94   0.350    -12077.23    4272.822
         Tp7 |  -5749.014   4681.705    -1.23   0.219    -14924.99     3426.96
         Tp8 |  -4626.267   5442.481    -0.85   0.395    -15293.33      6040.8
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

* part d
csdid income i.female i.white, ivar(id) time(year) gvar(group_,1) agg(group) method(reg)
gvar() does not contain a valid varname
r(198);

end of do-file
r(198);

You can see that the results are very similar across estimation strategies in this example.

Question 4

Finally, the data that we have contains a variable called occ_score which is roughly a variable that measures the occupation “quality”. Suppose that we (i) are interested in including a person’s occupation in the parallel trends assumption, (ii) are satisfied that occ_score sufficiently summarizes a person’s occupation, but (iii) are worried that a person’s occupation is a “bad control” (in the sense that it could be affected by the treatment).

  1. Repeat parts (a), (c), and (d) of Question 1 but including occ_score in the parallel trends assumption. Continue to allow for 1 year of anticipation effects.
Solutions:
* part a
csdid income i.female i.white occ_score, ivar(id) time(year) gvar(group_m1)
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : least squares
Treatment model: inverse probability
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
g1985        |
 t_1984_1985 |  -2195.225   2440.551    -0.90   0.368    -6978.618    2588.168
 t_1984_1986 |  -4943.573   3281.093    -1.51   0.132     -11374.4    1487.251
 t_1984_1987 |  -5648.525   3620.118    -1.56   0.119    -12743.83    1446.776
 t_1984_1988 |  -5486.878   5572.441    -0.98   0.325    -16408.66    5434.906
 t_1984_1989 |          0  (omitted)
 t_1984_1990 |  -9355.867   5613.212    -1.67   0.096    -20357.56    1645.826
 t_1984_1991 |  -9341.843   5703.704    -1.64   0.101     -20520.9    1837.211
 t_1984_1992 |  -10108.97   6080.289    -1.66   0.096    -22026.12    1808.174
 t_1984_1993 |  -5529.899   5706.746    -0.97   0.333    -16714.92    5655.118
-------------+----------------------------------------------------------------
g1986        |
 t_1984_1985 |    3820.89   4530.396     0.84   0.399    -5058.523     12700.3
 t_1985_1986 |  -8340.061   4312.966    -1.93   0.053    -16793.32    113.1959
 t_1985_1987 |  -1140.513   6498.254    -0.18   0.861    -13876.86    11595.83
 t_1985_1988 |  -3872.362   7461.828    -0.52   0.604    -18497.28    10752.55
 t_1985_1989 |          0  (omitted)
 t_1985_1990 |  -245.3064   6313.429    -0.04   0.969     -12619.4    12128.79
 t_1985_1991 |   1163.806   6794.528     0.17   0.864    -12153.22    14480.83
 t_1985_1992 |   357.4786   8779.844     0.04   0.968     -16850.7    17565.66
 t_1985_1993 |  -573.4507   6968.622    -0.08   0.934     -14231.7     13084.8
-------------+----------------------------------------------------------------
g1987        |
 t_1984_1985 |  -9335.567   4410.302    -2.12   0.034     -17979.6   -691.5335
 t_1985_1986 |  -3340.615   3042.829    -1.10   0.272     -9304.45    2623.219
 t_1986_1987 |  -3382.371    3539.52    -0.96   0.339     -10319.7    3554.961
 t_1986_1988 |  -4249.202   4235.181    -1.00   0.316       -12550    4051.599
 t_1986_1989 |          0  (omitted)
 t_1986_1990 |  -2636.246     6104.5    -0.43   0.666    -14600.85    9328.355
 t_1986_1991 |   3600.966   5437.356     0.66   0.508    -7056.057    14257.99
 t_1986_1992 |   10870.46   11785.57     0.92   0.356    -12228.83    33969.76
 t_1986_1993 |  -1193.181   7416.785    -0.16   0.872    -15729.81    13343.45
-------------+----------------------------------------------------------------
g1989        |
 t_1984_1985 |  -6306.913   3159.742    -2.00   0.046    -12499.89   -113.9316
 t_1985_1986 |   3619.346    2662.03     1.36   0.174    -1598.137    8836.829
 t_1986_1987 |   6300.986   2972.358     2.12   0.034     475.2719     12126.7
 t_1987_1988 |   1669.278   3289.866     0.51   0.612    -4778.741    8117.297
 t_1988_1989 |          0  (omitted)
 t_1988_1990 |  -3975.376   9196.928    -0.43   0.666    -22001.02    14050.27
 t_1988_1991 |  -21181.34   4461.373    -4.75   0.000    -29925.47   -12437.21
 t_1988_1992 |  -15120.42   3683.726    -4.10   0.000     -22340.4   -7900.454
 t_1988_1993 |  -16136.74   6008.476    -2.69   0.007    -27913.14   -4360.344
-------------+----------------------------------------------------------------
g1990        |
 t_1984_1985 |   275.2798    2901.55     0.09   0.924    -5411.654    5962.213
 t_1985_1986 |  -2972.748   3410.265    -0.87   0.383    -9656.745    3711.249
 t_1986_1987 |  -1061.871   2635.957    -0.40   0.687    -6228.253     4104.51
 t_1987_1988 |  -6533.743   3737.122    -1.75   0.080    -13858.37    790.8823
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1989_1991 |          0  (omitted)
 t_1989_1992 |          0  (omitted)
 t_1989_1993 |          0  (omitted)
-------------+----------------------------------------------------------------
g1991        |
 t_1984_1985 |  -11419.67   5874.196    -1.94   0.052    -22932.88    93.54378
 t_1985_1986 |   -3525.39   2388.987    -1.48   0.140    -8207.719    1156.939
 t_1986_1987 |   2689.547    3469.31     0.78   0.438    -4110.176    9489.271
 t_1987_1988 |  -7336.207   2611.752    -2.81   0.005    -12455.15   -2217.267
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |  -10337.31   7428.801    -1.39   0.164    -24897.49    4222.877
 t_1990_1992 |  -19895.18   6569.463    -3.03   0.002    -32771.09   -7019.269
 t_1990_1993 |  -19597.76    7799.03    -2.51   0.012    -34883.58   -4311.945
-------------+----------------------------------------------------------------
g1992        |
 t_1984_1985 |  -7566.207    4103.36    -1.84   0.065    -15608.64    476.2303
 t_1985_1986 |   50.10903   2689.611     0.02   0.985    -5221.431    5321.649
 t_1986_1987 |   1781.744    2885.26     0.62   0.537    -3873.261    7436.749
 t_1987_1988 |   4377.377   1624.927     2.69   0.007     1192.579    7562.176
 t_1988_1989 |          0  (omitted)
 t_1989_1990 |          0  (omitted)
 t_1990_1991 |   3464.896   5045.996     0.69   0.492    -6425.074    13354.87
 t_1991_1992 |  -4041.183   4569.095    -0.88   0.376    -12996.44    4914.079
 t_1991_1993 |  -27091.49     5672.5    -4.78   0.000    -38209.39   -15973.59
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details
* part c
csdid income i.female i.white occ_score, ivar(id) time(year) gvar(group_m1) agg(event)
csdid_plot
Units always treated found. These will be ignored
....x........x........x........x........xxxxx....x
x.......xx...
Difference-in-difference with Multiple Time Periods

                                                Number of obs     =     11,164
Outcome model  : least squares
Treatment model: inverse probability
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Pre_avg |  -2546.874   1087.689    -2.34   0.019    -4678.706   -415.0422
    Post_avg |  -5619.052   2561.075    -2.19   0.028    -10638.67   -599.4367
         Tm7 |  -7566.207    4103.36    -1.84   0.065    -15608.64    476.2303
         Tm6 |   -5684.78   3625.367    -1.57   0.117    -12790.37    1420.808
         Tm5 |  -412.9818   1669.441    -0.25   0.805    -3685.026    2859.062
         Tm4 |  -1783.231   1788.886    -1.00   0.319    -5289.383    1722.921
         Tm3 |   -349.233   1785.558    -0.20   0.845    -3848.862    3150.396
         Tm2 |  -2970.318   2555.139    -1.16   0.245    -7978.297    2037.662
         Tm1 |    938.631    1974.94     0.48   0.635     -2932.18    4809.442
         Tp0 |  -4998.251   1843.334    -2.71   0.007    -8611.119   -1385.384
         Tp1 |  -7485.133   2722.598    -2.75   0.006    -12821.33   -2148.939
         Tp2 |  -11342.87   3001.815    -3.78   0.000    -17226.32   -5459.425
         Tp3 |  -7450.486   3197.453    -2.33   0.020    -13717.38   -1183.593
         Tp4 |  -4153.519   3706.137    -1.12   0.262    -11417.41    3110.375
         Tp5 |   353.3305   5085.962     0.07   0.945    -9614.971    10321.63
         Tp6 |  -3889.916    4217.45    -0.92   0.356    -12155.97    4376.133
         Tp7 |  -6074.714   4703.608    -1.29   0.197    -15293.62    3144.189
         Tp8 |  -5529.899   5706.746    -0.97   0.333    -16714.92    5655.118
------------------------------------------------------------------------------
Control: Never Treated

See Callaway and Sant'Anna (2021) for details

* part d
csdid income i.female i.white occ_score, ivar(id) time(year) gvar(group_,1) agg(group)
gvar() does not contain a valid varname
r(198);

end of do-file
r(198);
  1. What additional assumptions (with respect to occupation) do you need to make in order to rationalize this approach?