Hi ,
My variable of interest is an indicator variable (LAW) that equals 1 if a firm is in a state having passed a certain law in that year and 0 otherwise.
gen LAW=state=="AZ" & fyear>2005|state=="AR" & fyear>2004|state=="CA" & fyear>1995|state=="CT" & fyear>2012|state=="DE" & fyear>1994|state=="FL" & fyear>1994|state=="GA" & fyear>1995|state=="HI" & fyear>2003|state=="IL" & fyear>2006|state=="IN" & fyear>1997|state=="KS" & fyear>2014|state=="LA" & fyear>1995|state=="ME" & fyear>1997|state=="MD" & fyear>2002|state=="MA" & fyear>1998|state=="MO" & fyear>2003|state=="NE" & fyear>1994|state=="NV" & fyear>1998|state=="NM" & fyear>2000|state=="NY" & fyear>1991|state=="OK" & fyear>2015|state=="OR" & fyear>2000|state=="PA" & fyear>1999|state=="RI" & fyear>1994|state=="TX" & fyear>2010|state=="TN" & fyear>1996|state=="UT" & fyear>2000|state=="VT" & fyear>2003|state=="WA" & fyear>2009 & fyear<2016|state=="MN" & fyear>1999 & fyear<2018
Thus, I would like to conduct placebo tests by randomizing years and locations to assure that our results are not driven by spurious correlation.
Specifically, for the placebo test based on time falsification, I want to randomly select years for each LAW adoption and repeat the same analysis 1000 times to get the average coefficients and t-stats of the variable of interest (LAW).
In addition, I want to have the placebo test based on location falsification by randomly selecting states of falsified LAW adoption and repeat the same analysis 1,000 times to produce the average coefficients and t-statistics of the variables of interest.
Anyone can shed light on how to conduct these analysis? Thank you in advance!
My variable of interest is an indicator variable (LAW) that equals 1 if a firm is in a state having passed a certain law in that year and 0 otherwise.
gen LAW=state=="AZ" & fyear>2005|state=="AR" & fyear>2004|state=="CA" & fyear>1995|state=="CT" & fyear>2012|state=="DE" & fyear>1994|state=="FL" & fyear>1994|state=="GA" & fyear>1995|state=="HI" & fyear>2003|state=="IL" & fyear>2006|state=="IN" & fyear>1997|state=="KS" & fyear>2014|state=="LA" & fyear>1995|state=="ME" & fyear>1997|state=="MD" & fyear>2002|state=="MA" & fyear>1998|state=="MO" & fyear>2003|state=="NE" & fyear>1994|state=="NV" & fyear>1998|state=="NM" & fyear>2000|state=="NY" & fyear>1991|state=="OK" & fyear>2015|state=="OR" & fyear>2000|state=="PA" & fyear>1999|state=="RI" & fyear>1994|state=="TX" & fyear>2010|state=="TN" & fyear>1996|state=="UT" & fyear>2000|state=="VT" & fyear>2003|state=="WA" & fyear>2009 & fyear<2016|state=="MN" & fyear>1999 & fyear<2018
Thus, I would like to conduct placebo tests by randomizing years and locations to assure that our results are not driven by spurious correlation.
Specifically, for the placebo test based on time falsification, I want to randomly select years for each LAW adoption and repeat the same analysis 1000 times to get the average coefficients and t-stats of the variable of interest (LAW).
In addition, I want to have the placebo test based on location falsification by randomly selecting states of falsified LAW adoption and repeat the same analysis 1,000 times to produce the average coefficients and t-statistics of the variables of interest.
Anyone can shed light on how to conduct these analysis? Thank you in advance!
Comment