maheshwari.anoop
New Member
P value in the case of two independent dummy variables (2010& 2011) is showing as a num# error.
Please help
Please help
Need Help on Regression Analysis
- Thread starter maheshwari.anoop
- Start date
maheshwari.anoop
New Member
We provide strategic consulting to transporters where one of the areas is cost optimization.
Repair & maintenance cost is roughly around 5% of the sales which is a controllable & significant cost for them. Below data covers the R&M cost for ~ 300 vehicles over the course of 3 years. Regression analysis is utilized to understand the projected cost based on various independent variables. This will assist the clients in scheduling their vehicles in specific routes, preventive maintenance and which factors are driving the cost
Let me explain the spreadsheet variables:
It consists of the following variables:
Dependent variable: Total cost or Cost per Km ( I am working to understand the predicted annual cost per vehicle)
Independent variables:
a) # of Km traveled
Dummy Variables:
a ) Year of Manufacturing ( 2009 - 2019) where 2019 is considered as a base case
b) Vehicle Type ( TR-08, TR-09, TK5) where TK5 is considered as a base case
TR means Trailors, TK means Truck
c) OEM ( Tata, Mahindra, AL) where Mahindra is considered as a base case
d) Data for 3 yrs is considered i.e. FY16-17, FY17-18 and FY18-19, where FY18-19 is considered as a base case
Repair & maintenance cost is roughly around 5% of the sales which is a controllable & significant cost for them. Below data covers the R&M cost for ~ 300 vehicles over the course of 3 years. Regression analysis is utilized to understand the projected cost based on various independent variables. This will assist the clients in scheduling their vehicles in specific routes, preventive maintenance and which factors are driving the cost
Let me explain the spreadsheet variables:
It consists of the following variables:
Dependent variable: Total cost or Cost per Km ( I am working to understand the predicted annual cost per vehicle)
Independent variables:
a) # of Km traveled
Dummy Variables:
a ) Year of Manufacturing ( 2009 - 2019) where 2019 is considered as a base case
b) Vehicle Type ( TR-08, TR-09, TK5) where TK5 is considered as a base case
TR means Trailors, TK means Truck
c) OEM ( Tata, Mahindra, AL) where Mahindra is considered as a base case
d) Data for 3 yrs is considered i.e. FY16-17, FY17-18 and FY18-19, where FY18-19 is considered as a base case
Hi:
One reason why you get error is because your sample size is too small for the no: of independent variables you are using in your model, try to see if you can drop some independent variable from the model. One way to do is to check for multicollinearity among the independent variable and drop the variable where there is high multicollinearity between the independent variable. I guess you will be familiar with this concept as you are into strategic consulting and may have come across these type of issues in your other projects.
Thanks
One reason why you get error is because your sample size is too small for the no: of independent variables you are using in your model, try to see if you can drop some independent variable from the model. One way to do is to check for multicollinearity among the independent variable and drop the variable where there is high multicollinearity between the independent variable. I guess you will be familiar with this concept as you are into strategic consulting and may have come across these type of issues in your other projects.
Thanks
p45cal
Well-Known Member
maheshswaro.anoop, the information you have given is very useful, but I was really after what data ranges you used (and on what sheets) as inputs to the regression analysis tool and importantly just which Excel regression tool you have been using.
Unless Nebu can help me with that as he seems to have understood better than me.
Unless Nebu can help me with that as he seems to have understood better than me.