2.reasoning the same as the above, resulting in 0.00217
3.R-spuared can be absolute coefficient (2 on the right side of the square) R2 = ESS/TSS = 1-RSS/TSS
Here R2 there are two algorithms, the first is in accordance with the above formula. R2=ESS/TSS=1-RSS/TSS
Here there are two algorithms for R2, the first one is according to the above formula
Based on the data given, we can know the RSS and the residual squared sum squared resid
TSS can be found out from the S.D. dependent var. TSS/(n-1)], where n=20, i.e. sample size, value of included observations.
Based on the formula, the value of R2 can be obtained.
The second method is based on the value of Adjusted R-squared below.
Adjusted R-squared = 1-(n-k)/(n-1)*(1-R2)
Where k=2, and the number of variables
This gives R2
4.S.E. of regression is an unbiased estimation of the variance of the following perturbation term, which in one-way linear regression is equal to RSS/(n-2)
5. F-value=[ESS/(n-k)]/[RSS/(n-1)]=[R2/(k-1)]/[(1-R2)/(n-k)]
You'll just have to figure out the answer for yourself, Mr. Owner
Here's the analysis
The first thing you need to do is to list the results of the regression analysis
What's included in this is the estimates that were derived, their corresponding t-value, se-value, and overall R2, F-value
Then the resulting data are analyzed
in terms of the following
1. t-value
The t-value is large, exceeding the critical value, and thus the significance of the two estimates is higher, respectively
2. Extremable coefficient vs. corrected extremable coefficient
The extremable coefficient 0<. = R2<=1, if R2 is close to 1, it means that the model overall fit is better, close to 0, it means that the overall fit is worse
Corrected extinction coefficient is generally used in multiple regression, is excluded from the increase in variables caused by an increase in the R2 of the extinction coefficient of the factors, the principle of judgment is the same as above. (The modified extinction coefficient may be greater than 1)
3. F value
F~F(k-1,n-k), when F is greater than the critical value, it means that the overall significance of the whole model is high.
For one-way linear regression, judging these is enough.
LZ to learn ah. These are very basic content ah.