$```$

load golfdata.txt $\%$ load data from file into matrix $`$golfdata$`$

driving $=$ golfdata$($:$,1)$;

scoring $=$ golfdata$($:$,2)$;

putting $=$ golfdata$($:$,3)$;

birdies $=$ golfdata$($:$,4)$;

$```$ You are given population data for a cubic model $\backslash ($ f $\backslash )$ that varies with time as follows:

$\backslash [$

f$($t$)=$a t$^\{3\}+$b t$^\{2\}+$c t$+$d$+\backslash$delta$\_\{$t$\}$

$\backslash ]$

where $\backslash ($ a$,$ b$,$ c $\backslash )$ and $\backslash ($ d $\backslash )$ are constants and $\backslash (\backslash$delta$\_\{$t$\}\backslash )$ denotes a time$-$varying noise parameter. You do not need to compute the function.

Find the regression model of order $3$ to fit the given data with time as the independent variable using polyfit.

$1.$ Predict the population as a function of time and store the results in $\backslash (\backslash$mathbf$\{$f$\}\backslash )\_$pred.

$2.$ Next, find out the mean absolute error between the prediction and the original data and store it in maevalue. not occur at the same time instance.

Script $($

$```$

rng$(101)$

time $=1$:$0\mathrm{.}5$:$50$;

delta $=$ randi$([-20,20],$ size$($time$))$;

a $=-0\mathrm{.}005$;

b $=0\mathrm{.}28$;

c $=0$;

d $=100$;

f $=$ a $*$ time.$^3+$ b $*$ time.$^2+$ c $*$ time $+$ d $+$ delta;

$\%$ Put your code here.

f$\_$pred $=$ zeros$(1,$ length$($time$))$; $\%$ placeholder, change it

$\%$ to compute predicted

$\%$ population

$```$