Hello Statalist,

I'm new to STATA and I have a homework where I need to find a subject and try to generate a regression model. So I decided to see if there is any impact on the stock market when the President is Democrat or Republican.

I have these following variables from 1990m1 to 2021m7 :

Dependant variable :
ln(S&P500 index)

Independant variables :
Recession : dummy variable = 1 if economy is in recession
Unemployment rate : for every month
10 years US Governement bonds : yield every month
Inflation : % every month
FED Funds Futures : for every month
Democrat President : dummy variable = 1 if Democrat President
Unified : dummy = 1 if Government is unified (House, Senate and White House)
Population majority : dummy = 1 if 50% or more of the population are satisfied with President work

I tried many regression and I'm not really confident about it. I know that every variable need to be stationary so I tried to use FD for ln(S&P500), government bonds and FED Funds (because these variables are non-stationary after a Dickey-Fuller test). This is the regression that I made :

regress d.lspnominal recession UN d.gouv_bonds inflation d.fed_funds_futures pres_democrat unified pop_majority

Nothing is significant under this regression. I don't know if I'm missing something or I just don't undertsand time series.

Also, I tried this ''normal regression'' :

reg lspnominal recession UN gouv_bonds inflation fed_funds_futures pres_democrat unified pop_majority time

Here, almost everything is significant but it can't be that easy to run a time series regression.

I upload my dta file here if anyone want to take a look. If you need more information about my problem, just ask

I'm a very desperate guy!

Thank you very much for your time!! And have a nice day