by

Rich Moushegian


Write-Up #12: Explorations with Excel

1. First we can investigate f(x) = (x) ^ 2 using the Excel Spreadsheet. If we have tabular values for x, we can construct a chart using Excel and the function as follows:

X Y
-3 9
-2 4
-1 1
0 0
1 1
2 4
3 9

By using the Graphing Wizard, we can construct a graph by connecting the data points.

2. If we investigate the following polar coordinate equation

we can place on spread sheet with 0.01 radians in the first column and the second column as the functional value. Then we increase the radians by .01 to obtain the following:

which produces the following graph as a "Radar Graph" (Chart Wizard icon):

3.Given the following data for predictions of the number of board feet (in 100's) depending on the age of the trees.

Age of Tree 100s Board Feet
20 1
40 6
60
80 33
100 56
120 88
140
160 182
180
200 320

Note that there are some missing data, but Graph Wizard can still plot the data.

Excel has the capacity to curve-fit the data with a power function as shown on the graph. As a result of the function, the missing data can be estimated and rounded as shown in the third column. Hence, the data for tree ages 60, 140, and 180 are estimated, below.

Age of Tree 100s Board Feet Adjust Data
20 1 1.026
40 6 5.77383840563218
60 15.8625165479802
80 33 32.4924073434241
100 56 56.6669837236975
120 88 89.2666737375281
140 131.085370383987
160 182 182.851763558389
180 245.243110365443
200 320 318.894743621071
220 404.406966679918
240 502.350242854558

In addition to estimating missing data (interpolation), we can project/predict the board feet for older trees (extrapolation) using the same prediction function for trees older than 200 years as was originally provided in the data.

4. The next example concerns an investigation of a certain technology company's value in the stock market. We have the following data: The first date is July '95 (Excel changed the date to "33419"), the second date is Jan '96 ("33603"), the third date is July '96 ("33785"), and so on.

We wish to interpolate the Dollar Value for July '97 ("34150") and extrapolate the value for January and July of 2000.

Actual Date Year-Date Dollar Value Calculated Val
33419 95.5 96 97.8938500000149
33603 96 99.6 96.3744000000079
33785 96.5 110.4 98.1266500000056
33969 97 111.3 103.150600000001
34150 97.5 111.446250000015
34334 98 121.2 123.013600000013
34515 98.5 138.1 137.852650000008
34699 99 152.4 155.963400000008
34880 99.5 189.3 177.345850000005
100 202.000000000015
100.5 229.925850000014

As a result of curve-fitting in Chart Wizard using the polynomial curve (second degree), we have the graph and formula as shown below:

The Calculated Value column is derived from the following formula, and data from the second column (Year-Date) is used to predict Dollar Value. The results are posted in the fourth column (Calculated Value).

We now have values for July '97 ($111.45) , Jan '00 ($202.00), and Jul '00 ($229.93).

 


RETURN to my Home Page.