paulina.94
New member
- Joined
- Apr 20, 2017
- Messages
- 1
Hi! I'm in the process of writing my thesis, in which I examine how the provision of non-audit services affects auditor independence. My study is based on data collected from the annual reports of Swedish listed companies during a time period of 4 years (from 2012 to 2015), so if I understood the terminology correctly, I'm working with panel data. The method I chose requires statistical analysis (linear and logistic regression) but since I'm a business administration student, my knowledge about statistics and SPSS is limited to two basic courses and some youtube tutorials.
I have a general idea of how regression models work and how to interpret the result but I have a problem with my last hypothesis. It's a pooled OLS with the natural logarithm of non-audit fees as the dependent variable and 10 independent variables (not counting the year dummies). 6 of them are nominal values, mostly financial ratios or auditor fees and the remaining 4 are dummy variables (SHORT TENURE, LOSS, OPINION and BIG 4). My equation looks like the one in the picture since I used it as inspiration. The only difference is that I have excluded the variables SQRTSUB and MINING and I have 4 years instead of 3 so I need to add 3 time dummies. I guess that the most logical thing to do would be to use 2012 as my base year and create dummies for 2013, 2014 and 2015.
(I'm not sure if it's important but as a side note, my pooled sample consists of 791 observations: 197 in 2012, 198 in 2013, 197 in 2014 and 199 in 2015.)
My question is this: how do I include the dummy variables for the different years in the equation? In the pooled SPSS file, I created a new variable that indicates which year the observation comes from but I'm not sure how to proceed from there. Should I create dummy variables for each of my 6 nominal variables and then include all the dummies in the regression model along with the variables for my base year (the ones from the equation)? And what about the variables that are already dummies, do I need to recode them into new dummies? Or do I only need to create dummies for the new variable that indicates the year? I'm confused because the equation I'm basing my own equation on only has two variables called 2009 and 2010. If they had dummy variables for each independent value, wouldn't those be included in the equation? I'm not sure if I'm making any sense but like I said, I'm not very good at statistics.
I'm sorry if it's a stupid question but I would really appreciate if someone could answer and explain this to me in a simple, uncomplicated way. I already tried asking a statistics teacher and I searched for tutorials on youtube but I didn't find the answer I was looking for. What I need is for someone to guide me through this step by step and tell me which variables should be recoded into dummies, how to do it (although I think I understand this part), and which variables should be included in the regression.
I have a general idea of how regression models work and how to interpret the result but I have a problem with my last hypothesis. It's a pooled OLS with the natural logarithm of non-audit fees as the dependent variable and 10 independent variables (not counting the year dummies). 6 of them are nominal values, mostly financial ratios or auditor fees and the remaining 4 are dummy variables (SHORT TENURE, LOSS, OPINION and BIG 4). My equation looks like the one in the picture since I used it as inspiration. The only difference is that I have excluded the variables SQRTSUB and MINING and I have 4 years instead of 3 so I need to add 3 time dummies. I guess that the most logical thing to do would be to use 2012 as my base year and create dummies for 2013, 2014 and 2015.
(I'm not sure if it's important but as a side note, my pooled sample consists of 791 observations: 197 in 2012, 198 in 2013, 197 in 2014 and 199 in 2015.)
My question is this: how do I include the dummy variables for the different years in the equation? In the pooled SPSS file, I created a new variable that indicates which year the observation comes from but I'm not sure how to proceed from there. Should I create dummy variables for each of my 6 nominal variables and then include all the dummies in the regression model along with the variables for my base year (the ones from the equation)? And what about the variables that are already dummies, do I need to recode them into new dummies? Or do I only need to create dummies for the new variable that indicates the year? I'm confused because the equation I'm basing my own equation on only has two variables called 2009 and 2010. If they had dummy variables for each independent value, wouldn't those be included in the equation? I'm not sure if I'm making any sense but like I said, I'm not very good at statistics.
I'm sorry if it's a stupid question but I would really appreciate if someone could answer and explain this to me in a simple, uncomplicated way. I already tried asking a statistics teacher and I searched for tutorials on youtube but I didn't find the answer I was looking for. What I need is for someone to guide me through this step by step and tell me which variables should be recoded into dummies, how to do it (although I think I understand this part), and which variables should be included in the regression.