cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

regression in x- and y-axis

regression in x- and y-axis

I must apologize for my bad English.

I have Mathcad PLUS 6.0. I'm trying to get a linear and a parabola regression for a few values for x- and y-axis. I tried to adapt the help, but it's using 3 dimensions. I think I know how to use regress, but I don't understand, how to handle interp.

I also have cases, in which I should force the curve through the origin and put simple units near the axis. I expect that those units can be found in the menus so I don't have to make them up.

I would appreciate it, if someone made it so easy for me, that all I have to do is to give the values and the units, decide if the regression was linear or parabola and if the regression should be forced through the origin.

You're my last hope. Please help me.

31 REPLIES 31

regression in x- and y-axis

Show something. The regress may not be the best fit ? A regress of n-1 degree than the 'n' points in the data, is then a Lagrange polynomial. It passes through all the points.

regression in x- and y-axis

What do you mean "show something"? Did you mean my values? Well, one of them all is about the swing of a pendulum. I need to get the speed on y-axis and the time on x-axis.

regression in x- and y-axis

See my article "Price of an Old Car or The way from Correlation to Regression in Mathcad":
http://twt.mpei.ac.ru/ochkov/car/c_e.html
Valery Ochkov
http://twt.mpei.ac.ru/ochkov

regression in x- and y-axis

On 12/23/2001 2:16:00 PM, kallekustaa6 wrote:
>What do you mean "show
>something"? Did you mean my
>values?

He means post a Mathcad worksheet (or a gif) that includes your values and as much else as you can.

Richard


regression in x- and y-axis

This is about the swing of the pendulum. The speed was measured at the lowest point of the swing. The values aren't good, but don't let that bother you.

So, I should put the height on the x-axis, and the velocity on the y-axis. Then I'd see that the regression is a parabola and would make another graph. This time the velocity2 would be on the y-axis. The regression should be linear and the derivate is what I really need to find. Please notice that in both cases the regression should be forced through the origin.

The kinetic energy of the pendulum at the lowest position = the potential energy at the highest position


� * m * v2 = g * m * h
� * v2 = g * h

v2 / h = 2 * g , where g = 9,81 m/s2

regression in x- and y-axis

I'm so sorry. I blundered. Again. So the .gif is right here.

regression in x- and y-axis

Try so:

Valery Ochkov
http://twt.mpei.ac.ru/ochkov

regression in x- and y-axis

Thank you for your answer, but that .mcd doesn't open with my Mathcad PLUS 6.0.
I don't want to be rude, but the regression described in the .gif is no use for me as it isn't forced through the origin (is that the right word for coordinates (0,0)).
And what about the non-linear regression? I'd still be glad if someone could help.

Thank you!

regression in x- and y-axis

Give us again a graph of what you want to regress. You can force regression near either ends or centered. If you go back may be a month or two you will see a sample *.gif I posted.