Table 1 Estimations examining the effect of EDR assignment selection (intent-to-treat test)

From: Incentive based emergency demand response effectively reduces peak load during heatwave without harm to vulnerable groups

 

EDR assignment selection

Heterogeneous effect

 

Main effect

Spillover effect

Urban

Children

Elderly

 

(1)

(2)

(3)

(4)

(5)

Treatment×Post×Indicator

  

−0.0284*** (0.0036)

−0.0041 (0.0733)

−0.0559** (0.0264)

Treatment×Post

−0.0155*** (0.0023)

−0.0024 (0.0016)

0.0063** (0.0027)

−0.0144 (0.0710)

−0.0289* (0.0165)

Indicator×Post

  

−0.0010 (0.0094)

0.1053 (0.0866)

−0.0327 (0.0391)

Indicator × Treatment

  

0.0113** (0.0044)

−0.0211 (0.0561)

−0.0034 (0.0224)

Treatment

−0.0315*** (0.0039)

−0.0226*** (0.0017)

−0.0401*** (0.0048)

0.0016 (0.0527)

−0.0165 (0.0166)

Post

−0.0355*** (0.0090)

−0.0372*** (0.0093)

−0.0437*** (0.0122)

−0.1436* (0.0838)

−0.0302 (0.0344)

Indicator

  

−0.1209*** (0.0245)

0.0486 (0.0756)

0.1540*** (0.0440)

Cluster in group

Yes

Yes

Yes

Yes

Yes

Controls

Yes

Yes

Yes

Yes

Yes

Observation

410,258

378,114

410,258

7,466

10,322

F

61.98

79.73

51.94

2.692

11.24

R2

0.0085

0.0082

0.0094

0.0077

0.0147

  1. This table reports the estimated coefficients and cluster-robust standard errors (in parentheses). The dependent variable in all columns is electricity usage during on-peak times. Columns (1) and (3–5) estimate the effect of assignment selection comparing the average outcome for households selected in the random assignment (EDR group and no-reply group, which we called assignment winners) to the average outcome for control households (no-notification group those not selected by the assignment) among heterogeneous groups. Column (2) compares the difference between the no-reply group and the no-notification group. The standard errors are clustered at the household-group level. Significance is at ***p < 0.01, **p < 0.05, *p < 0.1.
Back to article page