Fitting a function data works better when the years are shifted to smaller numbers. So let t be the number of years since 1988 and y be the number of polio cases. Then our data points are
(0,350), (4,138), (8,36), (12,5), (17,3.2), and (19,1.3).
Question 1
Question 4, about half-life, suggests fitting an exponential curve y=a·bt to the data. Take the logarithm of both sides and use the properties of logarithms to rearrange the function to a linear equation:
y = a·bt
ln(y) = ln(a·bt)
ln(y) = ln(a) + t·ln(b)
If we let Y=ln(y), A=ln(a), and B=ln(b) then we have the linear equation Y=A+t·B.
Now we can do linear regression and fit the line of least squares to the data. You can use a calculator (using STAT>CALC>LinReg(a+bx) on a TI-84), Excel (using =LINEST()), or similar software, or do the calculations manually by using the formulas
slope: B=(n∑tiyi – ∑ti∑yi)/(n∑ti2 – (∑ti)2)
Y-intercept: A = (∑yi – b∑ti)/n
where n is the number of data points (so n=6 in this question).
We find B=-0.2978 and A=5.8783. So b=eB=e-0.2978≈0.74 and a=eA=e5.8783≈357.20. Therefore the exponential curve y=357.2(0.74)t is a reasonable fit for the data.
Question 2
The year 2016 corresponds to t=28. There will be y=357.2(0.74)28 ≈ 0.08 cases in 2016. (Or we could reason that this implies there will be zero cases.)
Question 3
y=1 when 357.2(0.74)t=1. To solve for t, use a logarithm:
0.74t = 1/357.2
ln(0.74t) = ln(1/357.2)
t·ln(0.74) = ln(1/357.2)
t = ln(1/357.2)/ln(0.74) ≈ 19.5
There will be 1 case of polio about 19.5 years after 1988, so sometime in 2007.
Question 4
When t=0, y=357.2. The half-life is the time it takes for the number of polio cases to be cut in half. We want to solve for t when y=357.2/2.
357.2(0.74)t = 357.2/2
0.74t = 1/2
ln(0.74t) = ln(1/2)
t·ln(0.74) = ln(1/2)
t = ln(1/2)/ln(0.74) ≈ 2.3
The half-life is 2.3 years.