I have the following question,
Statement:
A given exercise has the p-value of 0.08 and my alpha is 5%.
The exercise was using a linear regression model to predict some future value outside the range of my sample.
Ho: model predict that event A is true Ha: model predicts that event A can’t be true
The conclusion said, “If the sample data use to create the linear model is reliable guide to predict the current event/behavior, there is a 92 percent chance that Ha, not chance alone, that Ha is true”.
My answer so far,
The question ask if is true, I believe is a misinterpretation of the p value because according the alpha is a 95% Conf. that the model predict that event A is true. The 92% is out the context but I can’t see why.
Thanks.
First, the p-value is higher than the significance level of $5$% so the observed data is not inconsistent with the null hypothesis.
Second, the p-value does not in itself support reasoning about the probabilities of hypotheses but serves only as a tool for deciding whether to reject the null hypothesis or not.
Hence, you can not draw that conclusion based on the p-value.