Philippa M. Rhodes
Write-up 12
Use the following exploration to generate a function to predict
data.
a. Monitor the weight of a fetus from 9 weeks until birth
(40 weeks).
b. Enter the data on a spread sheet and construct a function
that will model the data.
c. Using the function, predict the weight of the baby when (s)he
is 2 months old or 1 year old.
d. Calculate a measure of the error between your model and the
observed data.
Can you see the fetus sucking his/her thumb?
This data was found on the website Pregnancy
Week by Week .
The weight is grams. (It may be helpful to know that a paperclip
weighs one gram). The weight of the baby at birth is 3800 grams
which is 8.55 lbs or approximately 8 lbs 9 oz.
I used Microsoft Excel to graph the data.
Then I constructed a function that would hopefully model my data.
Using the speadsheet, it is easy to manipulate the function to
see how its graph changes and compares to the graph of the data.
So, I ended up the following function and graph.
The function closely models the data until about the 33rd week.
Then the babies rate of growth changes.
We will use the function to make predictions if the baby had continued
to grow at that rate ( first 33weeks).
Remember Week one after the baby is born is considered
week 41 when using the formula.
At the end of two months, we use 48 weeks, and the baby would
weigh 22.68 kilograms which is about 51 lbs.
The baby would weigh 22,662.85 kilograms at the end of one year.
That's almost 51,000 lbs!!
As we can see, this prediction is unreal. Thus, the measure of
error between my model and the observed data is huge. It is approximately
5846.45.
Return to Philippa's EMT Homepage