Population Modeling with Least-Squares Error Analysis
Consider the following results of measurements of a function
P(t):
T 0 1 2 3 4
P 100 101 108 140 230
Assume that the data set can be modeled by a function P = P0
+ B ta
a) Determine
P0. Use the Ln-test (logarithm) to show that the function P = P0
+ B ta can be used to model the data set.
b) Use a
linearized model to calculate the model parameters a and b such that the least-squares error becomes minimal.
c) Calculate the relative model error (not the least-squares
error).





