A few days ago
Anonymous

statistics help?

These data represent the population of a certain geographic region: 1950, 34 milion; 1960, 38 million; 1970, 43 million; 1980, 47 million; 1990, 50 million; 2000 55 million. predict the population in the year 2010.

How would I set up a problem like this?

Top 3 Answers
A few days ago
Merlyn

Favorite Answer

Normally population is modeled with exponential growth functions but if you look at a scatter plot of this data you will see it is linear in this range.

A easy, no thought way to get the solution is to open up Excel and in the A column put in the year and in column B you have the population in millions. In some other cell use the FORECAST Function

=FORECAST( 2010, B2:B7, A2:A7)

this uses a linear regression model to predict future values based on know y and x values. the result is 59 million.

To do this by hand look up simple linear regression.

0

A few days ago
CannonBooks
There are many methods of extrapolating a forecast, but I’ll assume you’ll probably want a fitted value from Ordinary Least Squares.

The easiest way to do this is list the values in a column in Excel, highlight all of them, and then drag the formula cross (in the bottom right corner of the cells you highlighted) down one more cell. You can only do this if the x values progress evenly (here, the years go by tens).

If you actually want to calculate the fitted line yourself, you can do that — it’s a little bit of a pain:

If x is the year, and y is population, the slope of the fitted line will be b = Cov(x,y)/Var(x) and the intercept can be calculated by taking y bar minus b*x bar (bar means the mean of the sample x’s or y’s). Then the equation is y=bx+a.

By the way, the answer is 59 million.

0

A few days ago
lovin_2beme
Make a graph and say based on the data in the graph, what would the projected population be in 2010?

Charts in either Word or Excel would work fine

0