Table 16.

Extreme temperature, mortality rate with OLS regression & interaction with SES

	(1) Interaction with SES
≤ 5th percentile	0.011
	(0.099)
≤ 5th percentile # Low SES	0.57***
	(0.094)
≤ 5th percentile # Medium SES	0.259*
	(0.109)
> 5th and ≤ 10th percentile	0.402**
	(0.123)
> 5th and ≤ 10th percentile # Low SES	−0.474**
	(0.170)
> 5th and ≤ 10th percentile # Medium SES	−0.533**
	(0.176)
> 10th and ≤ 25th percentile	0.001
	(0.064)
> 10th and ≤ 25th percentile # Low SES	0.219***
	(0.058)
> 10th and ≤ 25th percentile # Medium SES	0.046
	(0.060)
≥ 75th and < 90th percentile	0.185***
	(0.045)
≥ 75th and < 90th percentile # Low SES	−0.309***
	(0.045)
≥ 75th and < 90th percentile # Medium SES	−0.292***
	(0.039)
≥ 90th and < 95th percentile	−0.012
	(0.069)
≥ 90th and < 95th percentile # Low SES	0.110
	(0.103)
≥ 90th and < 95th percentile # Medium SES	0.111
	(0.100)
≥ 95th percentile	0.044
	(0.055)
≥ 95th percentile # Low SES	0.368***
	(0.071)
≥ 95th percentile # Medium SES	0.319***
	(0.057)
Observations	25,200
Province-Month FE	YES
Month-Year FE	YES

Results are obtained estimating an OLS regression on the mortality rate with month-year and province-month FE, an interaction with the three SES groups and adding the control variables used in Eq. (1) for individuals aged above 65. Constant present but not reported. Control variables are precipitation, solar radiation, wind speed, relative humidity, GDP per capita, PM2.5, gender, SES and population density Standard errors clustered at the provincial level. Significance levels: ***p < 0.001, **p < 0.01, *p < 0.05