Multiple linear is an enhanced form of linear regression in which more than one independent variable useful to predict the value for single dependent variable (Y). This predictive value calculates from linear transformation of the X variables. The calculations are complex but interrelationships among all the variables should be taken into account in the weights assigned to the variables, Lesson 5: Multiple Linear Regression.
Hence, it’s very much useful when we want to calculate the value of one dependent variable by using the values from two or more other independent variables. In many businesses, it is very useful like; predicting future prices based on the demand and supply, estimate the demand of the particular time using the historic details for that particular time, estimate the probability of a prospect becoming a customer given attributes, and many more. Hence, multiple regression is very much helpful to predict the future values for the business situation including; any business type such as; a shop, retail, mall, etc. Lesson 5: Multiple Linear Regression.
Multiple regression wouldn't work if there is no statistical data and if there is single independent variable then it doesn't work like; if a nation wants to calculate the censuses based on the individual age only, business want to know the prices based on supply, etc. Also, it wouldn't work for theoretical data as well, Lesson 5: Multiple Linear Regression.
In my professional experience, I wouldn't prefer to use multiple regression in calculating the marks of each student based on few other students’ marks because it can't define the marks for any student because here, all variables are independent and couldn't predict. In addition to this, I would prefer to use multiple regression in letting understand the share market prediction based on some independent variable value, Lesson 5: Multiple Linear Regression.