+ Reply to Thread
Results 1 to 1 of 1

Thread: Durbin-Watson statistic below 2, fixed effect model

  1. #1
    Points: 6, Level: 1
    Level completed: 11%, Points required for next Level: 44

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Durbin-Watson statistic below 2, fixed effect model




    Hello

    I am asking for your advice. I am working with unbalanced panel data set. Sample contains data about 11 largest Finnish insurance companies, time period is 13 years. Dependent variable is profitability indicator ratio; independent variables are concentration measured by HHI index and artificial continuous variable to consider law change and control variable is market growth.

    Please see attached picture (regression I) about regression analysis output.

    P-value of market growth showed that variable is insignificant, so I run regression again. Please see attached picture (regression II).

    Durbin-Watson statistic value makes me concerned - it should be around 2. In my model it is 1,56 and it indicates positive autocorrelation. Please see attached picture (correlation)

    Could you please tell me if I have any point to deal more with that econometric model or the results are meaningless and I should try to find another way for composing my school project empirical part? Please give some advise how I could use/fix that model. I planned to estimate same relationship for other Nordic countries.

    Now I have one quite silly question. I am confused about the plus-minus sign () of regulation dummy variable. Dependent variable is combined ratio and lower value indicates higher profitability. So, I do not understand how to read the output in this case. My intention was to prove that changes in the law (regulation dummy) affected positively the insurers` profitability. What sign (+/-) in front of regulation dummy I am looking for?

    Thank you in advance
    Attached Thumbnails Attached Thumbnails Click image for larger version

Name:	RegressionI.jpg‎
Views:	14
Size:	25.6 KB
ID:	5220   Click image for larger version

Name:	RegressionII.jpg‎
Views:	18
Size:	24.9 KB
ID:	5221   Click image for larger version

Name:	Correlation.PNG‎
Views:	5
Size:	1.7 KB
ID:	5222  

+ Reply to Thread

           




Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats