[Gabriel Temesgen]

Dear Khati Zolfaghari greetings. I am running a local projection method however the "lp" shows unrecognized. How can I define an ado file in STATA for LP method? Thanks

[Khati Zolfaghari]

@Gabriel Temesgen hello, you can look at this also, it is so helpful. https://sites.google.com/site/oscarj...al-projections

[Gabriel Temesgen]

Dear Khati Zolfaghari,

I have four questions. I am running a panel local projection method. 1. what is the role of leads in local projection (LP) models? Do they have a relationship with lags? 2. in Estimate the LP, I saw two papers use different commands in the highlighted one. the command of Jorda uses if _Year==`h' while another paper use _Year==`h'+2. What does the _Year indicate? when I try both they gave me different IRFs. May you give me what it indicates. qui forvalues h = 0/5 { xtreg F`h'.drgdp l1drgdp l2drgdp F0dggcbp_gdp l1dggcbp_gdp l2dggcbp_gdp `controls' tdum* , fe replace LPCAPB = _b[F0dggcbp_gdp] if _Year==`h' replace LPCAPBse = _se[F0dggcbp_gdp] if _Year==`h' } or qui forvalues h = 0/5 { xtreg F`h'.drgdp l1drgdp l2drgdp F0dggcbp_gdp l1dggcbp_gdp l2dggcbp_gdp `controls' tdum* , fe replace LPCAPB = _b[F0dggcbp_gdp] if _Year==`h'+2 replace LPCAPBse = _se[F0dggcbp_gdp] if _Year==`h'+2 } 3) when I transform my variables to log form and multiply by 100, I got IRFs 100 percent higher than the IRfs with log form only as depicted below. what is the reason? I couldn't know the justification in the literature they use either one of them. I couldn't get the reason why some authors multiple by 100 while transforming their variables to logarithmic form. can u tell me the reason why? gen lg = 100*ln(spending) gen lg = ln(spending) 4) I am confused with the difference between "one standard error innovation" and "percentage innovations (impulses)". What is their difference? Does the default result of STATA in percentage change? Thanks in advance. Best