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

Community Tip - Did you get an answer that solved your problem? Please mark it as an Accepted Solution so others with the same problem can find the answer easily. X

Jackknife in Mathcad 15

DickCun
1-Newbie

Jackknife in Mathcad 15

Is anyone aware of how to create a matrix of a data vector using the Jackknife procedure in Mathcad 15?

1 ACCEPTED SOLUTION

Accepted Solutions

Thanks so much.....your solutiom is very elgant!!!

View solution in original post

8 REPLIES 8

Which problems have you encountered when you tried to implement it in Mathcad?

Attach your sheet to get help with your approach.

I have used this process in Excel, as well as the Bootstrap in Excel and Mathcad. At the present I am stumped as to how to accomplish the Jackknife in Mathcad an thus have not provided a Mathcad sheet. I am currently analyzing data found in Bury's text examples using both Log-Normal and Gumbel distributions. Ultimately, I would use this method or the Bootstrap in the analysis of snow loads. I would like to find the standard error of the 50 yr, 100 yr mean recurrence interval.

MikeArmstrong
5-Regular Member
(To:DickCun)

I suggest you post what you have attempted thus far to give collabs on here something to work with.

If Wikipedia is right.

Not sure about variance and bias.

Thanks for your help Fred. It helped me develope the attached Mathcad 15 spreadsheet. May be you can help further. Would appreciate it.

MikeArmstrong
5-Regular Member
(To:DickCun)

I think you need to revisit the basics of your program. You are using submatrx and trying to pass a range to the lower indices to be extracted for a single column vector, obviously this is not going to work.

You gain more flexibility by creating utility functions instead of static calculation routines:

1.png

Initializing new and JM would probably mean better programming style but isn't absolutely necessary as both variables are treated as metrices from the very first moment and so are local anyway.

BTW, your way of deleting the vector element using submatrix and stack will fail if by chance the very first (index) or the very last element should be deleted. My brute force way should suffice as long as the magnitude of the data vector is within a reasonable range. BTW, as unbelievable as it may sound, stack is much slower than an equivalent routine written using loops.

I guess the attached sheet will help

2.png

Thanks so much.....your solutiom is very elgant!!!

Top Tags