Can Matlab help in design experiments?

1 次查看（过去 30 天）

Ayob 2015-1-8

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/169257-can-matlab-help-in-design-experiments

评论： dpb 2015-1-9

I have four different type of inputs and three type of outputs, I want to design my experiments with these values in Matlab and I don't know how can I do it? My inputs are Continuous values as given below:

I1=[1.5,5.5]mm

I2=[5,85]degree

I3=[100,300]mm

I4=[5,50]mm

And Outputs ranges are unkown:

O1=[?,?]m/s

O2=[?,?]pa

O3=[?,?]kg

I need two methods to design my experiments with continuous values and discontinuous values whether I prefer to write Input ranges in discrete numbers like:

I1=[1.5, 2.5, ... 5.5]mm

I2=[5, 10, 15,... 85]degree

I3=[100, 150, 200, .... 300]mm

I4=[5, 10, 15, ... 50]mm

7 个评论
显示 5更早的评论隐藏 5更早的评论

Ayob 2015-1-9

编辑：Ayob 2015-1-9

Yes, exactly. I need help to find the appropriate codess in Matlab.

(Are there codes for working with continuous values which are introduced in I1,I2..I4 ranges?)

dpb 2015-1-9

Yes, but still it's the objective of the experiement that is the key, first before one can decide what is the appropriate design. Hence the tool selection follows that.

The tool can't make the decision as to what it is that is the purpose of running the experiment in the first place; only your (apparently a well-kept secret as you don't seem willing to share even a hint) knowledge of what you're trying to accomplish as the end result can define that.

First, you need to write the null hypothesis of the test and decide the proper analysis technique that will let you compute and estimate the necessary statistics to test that hypothesis or if it is a model you're trying to build have enough background information on the process to be able to select an appropriate one and then be able to estimate the underlying coefficients.

You're putting the cart before the horse by, it seems, expecting one of these tools to do that for you magically.

请先登录，再进行评论。

请先登录，再回答此问题。