Skip to main content
. 2022 Oct 13;112:102365. doi: 10.1016/j.foodpol.2022.102365

Table D3.

Robustness Check for Regional Spillovers and Pre-trends with 1500 Clusters.

Variables Impact of Infections in Neighboring Cities Test for Pre-existing Trends from January 1–22, 2019
NRF NPI NRF NPI
(1) (2) (3) (4) (5) (6) (7) (8)
National −0.017*** −0.110*** −0.039*** −0.070*** −2.330 −1.428 −0.172 0.151
(0.003) (0.001) (0.009) (0.018) (1.716) (2.050) (0.409) (0.461)
Local −0.332*** −0.334*** −0.012*** −0.010*** 0.492 0.491 0.019 0.006
(0.064) (0.049) (0.003) (0.002) (0.432) (0.491) (0.098) (0.098)
Neighbor 0.098 0.043 0.015 −0.009
(0.091) (0.111) (0.023) (0.026)
K-means 1500 Yes Yes Yes Yes Yes Yes Yes Yes
Other Control Variables No Yes No Yes No Yes No Yes
City-month Yes Yes Yes Yes Yes Yes Yes Yes
Restaurant FE Yes Yes Yes Yes Yes Yes Yes Yes
City FE Yes Yes Yes Yes Yes Yes Yes Yes
Day of the week FE Yes Yes Yes Yes Yes Yes Yes Yes
Hour FE Yes Yes Yes Yes Yes Yes Yes Yes
Month FE Yes Yes Yes Yes Yes Yes Yes Yes
Observations 144,044 144,044 144,044 144,044 118,207 118,207 118,207 118,207
R-squared 0.127 0.203 0.142 0.277 0.501 0.561 0.237 0.365
Variables Test for Pre-existing Trends from January 23– March 18, 2019 Test for Pre-existing Trends from March 19-May 31, 2019
NRF NPI NRF NPI
(1) (2) (3) (4) (5) (6) (7) (8)
National −0.760 −2.543 0.529 0.730 −0.122 −0.286 1.214 1.795
(0.439) (1.517) (0.781) (0.531) (3.793) (5.828) (1.417) (1.220)
Local 0.016 0.010 −0.007 −0.009 0.129 0.071 0.072 0.077
(0.068) (0.053) (0.014) (0.008) (0.105) (0.111) (0.059) (0.056)
K-means 1500 Yes Yes Yes Yes Yes Yes Yes Yes
Other Control Variables No Yes No Yes No Yes No Yes
City-month Yes Yes Yes Yes Yes Yes Yes Yes
Restaurant FE Yes Yes Yes Yes Yes Yes Yes Yes
City FE Yes Yes Yes Yes Yes Yes Yes Yes
Day of the week FE Yes Yes Yes Yes Yes Yes Yes Yes
Hour FE Yes Yes Yes Yes Yes Yes Yes Yes
Month FE Yes Yes Yes Yes Yes Yes Yes Yes
Observations 188,251 188,251 188,251 188,251 271,641 271,641 271,641 271,641
R-squared 0.499 0.555 0.350 0.454 0.209 0.299 0.220 0.350

The regressions are estimated using weighted least squares, using as weight the daily number of transactions in each restaurant. Note that *, **, *** indicate 90%, 95%, and 99% levels of confidence. All errors are clustered at the city level.