
Economics Economics Forum  Financial Mathematics, Econometrics, Operations Research, Mathematical Finance, Computational Finance 
 LinkBack  Thread Tools  Display Modes 
June 4th, 2017, 06:37 AM  #1 
Newbie Joined: May 2015 From: sweden Posts: 12 Thanks: 0  Econometrics (regression)
Hello! I need help to interpret my regression results. i have a hard time understanding when there is percentage point or not. My regression: i have an excel file with following variables: Real GDP growth / capita i have calculated this myself with formula ((year 2 year 1)/year 1)* 100 Real GDP / capita year t1 just lagged one year back Delta minimum wage calculated myself with ((year 2 year 1)/year 1)* 100 , (thus i have the level of minimum wage too but its not included in my regression) Delta unemployment calculated myself as year 2  year 1 ex: 5,7  5,5 = 0,2 Unemployment year t1 expressed as 5,7 meaning 5,7% Regression 1: Y Real GDP growth / capita = delta minimum wage + Real GDP / capita t1 Regression 2: Y delta unemployment = delta minimum wage + unemployment t1 Results: regression 1: Y (Real GDP growth / capita) = 0,159*** + 0,0143*(minlwage) + 0,0002397***(gdp level) regression 2: Y (delta unemployment) = 0,890209** + 0,006508*(minwage) + 0,230477***(unemployment level) To interpret these results more i do this: regression 1: an increase of minimum wage by 10% will lead to a decrease in GDP growth/ capita by 0,143 %. I then took the average growth in GDP/capita over the period i studied and found 1,77%. Can i say that the new growth would be 1,77  0,143 = 1,627% ? Regression 2: an increase of the minimum wage by 10% will lead to an increase in unemployment by 0,06508%. I looked up the unemployment numbers in USA which is 4,4% and 7 000 000 people. Thus 1% means 1 590 909 people. New unemployment will be 4,4 + 0,06508 = 4,46508 . 4,46508 * 1 590 909 = 7103535 people. increased unemployment number is thus: 103535 people ? Am i interpreting this right? or do i mix percentage and percantage point? 
June 4th, 2017, 09:02 AM  #2 
Senior Member Joined: May 2016 From: USA Posts: 1,300 Thanks: 549 
It is debatable whether this question belongs in the statistics forum because, although your statistics are about economics, your question is about interpreting statistics. You have made this really confusing. If I understand you correctly, your data on the change in real GDP are NOT in percentage terms, but your data on the change in the minimum wage are. If my understanding is correct, your interpretation 1 is wrong. And it appears that the change in unemployment is in percentage points. As a further statistical comment, have you checked whether your overall model and each of the individual coefficients are statistically significant? About your economics, I see no reason to start with matching inflation adjusted figures (changes in real gdp) against nonadjusted numbers (changes in money minimum wages). That would be an interesting finding, but would require an explanation. 
June 4th, 2017, 09:30 AM  #3  
Newbie Joined: May 2015 From: sweden Posts: 12 Thanks: 0  Quote:
I know it can be hard to understand my numbers. i will try to explain: I have an column with Real GDP / capita numbers. ex: year 1 20000 , year 2 21000 Thus, my Real GDP growth /cap will be ((21000  20000) / 20000) * 100 = 5 Minimum wage is calculated in exaclty the same way. min wage year 1 10 , year 2 11 , thus delta min wage is 1110/10 *100 = 10 Concerning unemployment, I have an column with unemployment numbers like: 5,7 for 5,7% Delta unemployment is calculated by year 2  year 1 ex: 5,85,7 = 0,1 Furthermore, the findings are significant. Represented by the stars after each coefficient. Regarding the nominal minimum wages it is because trouble of finding infaltion numbers for all years. i found that an average increase in the minimum wage was big enough to make the effect neglectable. But it would ofcourse have been nice with Real numbers on minimum wage also.  
June 5th, 2017, 01:04 PM  #4 
Newbie Joined: May 2015 From: sweden Posts: 12 Thanks: 0 
ANYONE HELP: I just want to know if my reasoning behind this is correct or not? Regression 2: an increase of the minimum wage by 10% will lead to an increase in unemployment by 0,06508%. I looked up the unemployment numbers in USA which is 4,4% and 7 000 000 people. Thus 1% unemployment means 1 590 909 people. New unemployment will be 4,4 + 0,06508 = 4,46508 . 4,46508 * 1 590 909 = 7103535 people. increased unemployment number is thus: 103535 people ? Or else i will just interpret it as a 10% increase in minimum wage will lead to a 0,06508 % UNIT increase. And not evaluate it any further 

Tags 
econometrics, regression 
Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Baysian econometrics  canfly  Probability and Statistics  24  March 27th, 2017 09:05 AM 
Applied Statistics and Econometrics  sarwar  Probability and Statistics  0  September 19th, 2014 10:27 AM 
Linear algebra & Econometrics  BartL  Math Books  0  October 26th, 2010 12:17 AM 
Econometrics  princessanna57  Economics  1  September 11th, 2009 07:29 PM 