Question

Question 1 2 pts In regression analysis, an estimated coefficient is statistically significant when: the associated p-value is close to 1. 0 the ratio of the standard error divided by the estimated coefficient is close to 0. the associated p-value is close to O the ratio of the estimated coefficient divided by the standard error is close to 0. Question 2 2 pts You are reading an academic paper where one of the estimated coefficients has a p-value of 0.0965 O The coefficient is statistically significant at the 10% level. O The coefficient is not statistically significant at the 10% level. Question 3 2 pts Every possible data sample that you could use in regression analysis is always representative of the entire population True False

Answer 1

Q 1). In regression analysis an estimated coefficient is statistically significant when the associated p-value is close to zero (3^rd option ). P-value is defined as probability that null hypothesis is being accepted, i.e alternative hypothesis is insignificant. Probability is always between 0 - 1 and the alpha value ( 10% = 0.1, 5% =0.05, or 1% =0.01....) is taken as the cutoff for significance.

Q 2). If the estimated coefficient has p- value of 0.0965 , the coefficient is statistically significant at 10% level ( 1^st option). At 10% alpha value critical/cutoff the p - value will be 0.1. Since 0.0965< 0.1, means the estimated coefficient is statistically significant.

Q 3). False (2^nd option) . It is not necessary that every sample data that we take to get an inference about population always representative of the entire population. There can be biased samples in which one or more parts of population are favored over others.