Page 1 of 1

Why Thetas>1

Posted: Sat Mar 23, 2019 2:40 pm
by aliko
Could you please let me know where am I wrong? I got Ts>1 and big alpha value
but the fitting with these numbers work well

60 0.507
100 0.44
300 0.408
500 0.324
1000 0.313
2000 0.286
15000 0.269

Re: Why Thetas>1

Posted: Mon Apr 01, 2019 1:22 am
by Jirka
If you look at the results, you should see that the parameters thetas and alpha are fully negatively correlated (-1) (see the Correlation Matrix). This is also reflected in the 95% confidence limits of these estimated parameters, which both go from large negative to large positive numbers. This means that you cannot estimate these two parameters simultaneously from your data. You need to fix one and estimate only the other one. You would need to have more data closer to full saturation to be able to estimate these two parameters simultaneously. J.
Retc.png (39.98 KiB) Viewed 998 times