Question

What are the assumptions of regression? How does a correlation compare to regression model with only one predictor? (8 points)

Answer 1

sol::-

There are five main key assumptions for regression they are:-

1)linear relationship

2)multivariate normality

3)no or little multicollinearity

4)no auto correlation

5)Homoscedasticity

linear regression is an examination that surveys whether at least one indicator factors clarify the needy (paradigm) variable. The relapse has five key suspicions.

comprision of correlation with regression model:-

(1) Regression analysis considers the relation between independent variables and response variable, simultaneously; while correlation can be used pairwise.

(2) Regression analysis gives a predictor function which can be used to predict the value of response variable for a new sample in the terms of its values of independent variables.

The level of affiliation is estimated by a connection coefficient, meant by r. It is at times called Pearson's relationship coefficient after its originator and is a proportion of straight affiliation. On the off chance that a bended line is expected to express the relationship, other and more confused proportions of the connection must be utilized.

The estimation of a relationship coefficient can fluctuate from less one to in addition to one. A less one shows an ideal negative relationship, while an in addition to one demonstrates an ideal positive connection. A connection of zero means there is no connection between the two factors. At the point when there is a negative relationship between's two factors, as the estimation of one variable builds, the estimation of the other variable reductions, and tight clamp versa. As such, for a negative connection, the factors work inverse one another. At the point when there is a positive connection between's two factors, as the estimation of one variable builds, the estimation of the other variable additionally increments. The factors move together.

The standard blunder of a connection coefficient is utilized to decide the certainty interims around a genuine relationship of zero. In the event that your connection coefficient falls outside of this range, it is fundamentally not the same as zero. The standard mistake can be determined for interim or proportion type information (i.e., just for Pearson's item minute relationship)