Finding P-Values For a Null Hypothesis with Neutral Trend

#1
Please help me. I am so lost. I have data that is days above a certain temperatures every year from 1948-2019. So basically something like:
1948 36
1949 31
1950 36
1951 34
1952 36
1953 35
and so on. To find the trend I already did a least-squares linear regression and go that done. Now I am trying to find if the trend is significant or not by using p-values.
Here is where I am stuck. I need to calculate a p-value with the null hypothesis being a coefficient equal to zero (a neutral trend). I don't know how to do this. I am using excel if that helps I just do not know how to go about this. If someone could do a sample problem with the small data given that would be great. I also don't know how to correctly go about that null hypothesis. Thank you in advance for the help!
 

obh

Active Member
#2
Hi Claire,

You can do the t-test on the coefficient, or If you use only one IV the significance level of the all model (F test) and the significance level of the one coefficient are the same.

If you use the "data analysis - regression" in excel, you will get the "Significance F" in the ANOVA table
or the P-value of your IV

If you try to calculate by yourself:
F(p,n-p-1) = MS-Regression/ MS-residuals

You may see the F calculation based on the ANOVA table, and results:
http://www.statskingdom.com/410multi_linear_regression.html
 
Last edited:

katxt

Active Member
#3
You can also load the standard Excel Analysis ToolPak and do the regression from it. The p value is sitting there either in the anova table or the summary table. Plenty of examples on the net.
 

Attachments