Sunday

August 2, 2015

August 2, 2015

Posted by **grant** on Saturday, May 12, 2007 at 1:05pm.

The table shows the depth (d metres) of water in a harbour at certain times (t hours) after midnight on a particular day.

time t (hours)

0

1

2

3

4

5

6

7

8

9

10

11

12

13

and the in the next column corresponding to the figures above is

depth d (m)

3.0

3.3

4.2

5.6

7.2

8.2

9.0

8.9

8.1

7.3

5.6

4.3

3.5

3.1

Use the regression facilities on your calculator to fit a sine curve to these data. Choose the one option which provides the best fit model (with coefficient rounded to 2 significant figures).

A) t=3.0sin(0.49d-1.6)+6.0

B) d=3.0+6.0sin(0.49t-1.6)

C) t=2.9sin(0.48d-1.5)+6.0

D) d=3.0sin(0.49t-1.6)+6.0

E) d=2.99sin(0.49t-1.60)+6.04

F) d=6.0-3.0sin(0.49t+1.6)

Can anyone help me please. I may be d?

For Further Reading

math - Count Iblis, Saturday, May 12, 2007 at 9:58am

Note that you want d(t) to be of the form:

d(t) = A + B sin(r t + s)

Now, as you can see, all the options that express d like this have r = 0.49. This makes it easy to find the best fit, because you can now transform this problem into a linear regression problem as follows. You write:

d(t) = A + B sin(r t + s) =

A + B [cos(s) sin(r t) cos(s) + sin(s)cos(rt)(rt)] =

A + B cos(s) sin(rt) + B sin(s) cos(rt)

So, this problem is of the form:

d(t) = A_1 + A_2 f_2(t) +

A_3f_3(t)

with

f_2(t) = sin(rt)

f_3(t) = cos(rt)

So this is a standard linear regression problem as d is a linear function of f_2 and f_3.

You just need to proceed in the same way as you would proceed in case of a regression problem like:

Y = A + B X1 + C X2

for given data for Y, X1, and X2.

math - bobpursley, Saturday, May 12, 2007 at 10:37am

math - Count Iblis, Saturday, May 12, 2007 at 11:51am

I've just emailed you :)

math - grant, Saturday, May 12, 2007 at 11:04am

i'm not sure what you mean. which are the correct answers?

math - Count Iblis, Saturday, May 12, 2007 at 11:49am

Hi Grant,

Do you have a calculator in which you can enter data to do linear regression in multiple variables?

If so, then you need to do the following. Compile a table with the values of X1 = sin(0.49 t) and

X2 = cos(0.49 t).

Then you must enter the data for Y = d, X1 and X2 in your calculator and out will come the regression parameters corresponding to a fit of the form:

Y = A + B X1 + C X2

From B and C you then calculate the amplitude and the phase of the sin...

math - grant, Saturday, May 12, 2007 at 12:50pm

I don't have this type of calculator as yet and my homework is due on monday. Is the answer A?

Grant, the solution is posted here:

http://www.jiskha.com/display.cgi?id=1179090779