Scribd
Upload
7 views2 pages

Logistic Model for U.S. Population Growth

The document discusses the logistic model for population growth, detailing its parameters and their significance. It provides a step-by-step approach to fitting the model to U.S. Census data, including obtaining initial values for the parameters and using them in non-linear least squares estimation. The tutorial also includes tasks such as creating a scatterplot, fitting the model, and superimposing the fitted curve onto the scatter diagram.

Uploaded by

kiptooi277
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views2 pages

Logistic Model for U.S. Population Growth

The document discusses the logistic model for population growth, detailing its parameters and their significance. It provides a step-by-step approach to fitting the model to U.S. Census data, including obtaining initial values for the parameters and using them in non-linear least squares estimation. The tutorial also includes tasks such as creating a scatterplot, fitting the model, and superimposing the fitted curve onto the scatter diagram.

Uploaded by

kiptooi277
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

DEDAN KIMATHI UNIVERSITY OF TECHNOLOGY

SAS3202: REGRESSION ANALYSIS

Tutorial 7

ˆ A simple model for population growth towards an asymptote is the logistic model

β1
yi = + ei
1 + exp(β2 + β3 xi )

where is yi the population size at time xi , β1 is the asymptote towards which the
population grows, β2 reflects the size of the population at time x = 0 (relative to
its asymptotic size) and β3 controls the growth rate of the population.

ˆ To fit the logistic model to the U.S. Census data, the starting values for the
parameters are required. It is often important in non linear least squares estimation
to choose reasonable starting(initial) values, which generally requires some insight
into the structure of the model.

(a) Obtain a scatterplot of the data. What comments can you make from this plot?

(b) The intial values are

– β1 represents asymptotic population. The data in the plot shows thatin 1990
the population stood at about 250 million and did not appear to be close to
an asymptote; so as not to extrapolate too far beyond the data, we set the
starting value of to 350, i.e β1 = 350
– It is convenient to scale time so that x1 = 0 in 1790, and so that the unit of
time is 10 years.
– Then substituting β1 = 350 and x1 = 0 into the model, using the value
y1 = 3.929 from the data, and assuming that the error is 0, we have
350
3.929 = =⇒ β2 = 4.5
1 + exp(β2 + β3 (0))

– At time x = 1 (i.e., at the second Census performed in 1800), the population


was y2 = 5.308 . Using this value, along with the previously determined start
values for β1 = 350 and β2 ≈ 4.5, and again setting the error to 0.
– At time x = 1 (i.e., at the second Census performed in 1800), the population
was y2 = 5.308 . Using this value, along with the previously determined start
values for β1 = 350 and β2 = 4.5, and again setting the error to 0. Then
350
5.308 = =⇒ β3 ≈ −0.3
1 + exp(4.5 + β3 (1))

(b) With these starting(initial) values for the non-linear least squares algorithm, fit
the data and give the model summary. Give the fitted model.

(c) Superimpose the fitted curve onto the scatter diagram.

You might also like