Dear, Please help urgently. I used the Fixed Nonlinear Regression option. I checked the SQRT(x) option in the menu. This resulted in the table shown in the picture. What is the mathematical form of the equation? Best regards. ZB

[Zlatko Brievac]

[Fredrik Rosell]

Hello Zlatko,

To ensure that any responses are relevant for the particular product that you are using, please add a note about what that product is (and to make it even more clear, edit the question, and select a product in the Product field)

Thanks!

/Fredrik

[Zlatko Brievac]

Hello Frederik, I use Tibico Statistica 14. I performed the analysis using the fixed non-linear regression option (Fig1) and turned on the option (Fig 2), I also sent the intermediate step fig 3. As a result, I got the table in Fig 4. The problem is that I do not know how to write a mathematical equation based on toga. Usually, the mathematical expression is written according to column b of the table in fig 4, but I don't know how to do it. I looked at the examples provided by Statistica, but they are too simple and I couldn't do it based on them. I hope it's clearer now.

Kind regards,

Zlatko

[Fredrik Rosell]

Hello Zlatko,

There doesn't appear to have been any answers from Statistica gurus here yet. I reached out to a Statistica Support engineer, who didn't have any quick answer for you but he recommended that you open a support case for this (https://support.tibco.com)

[Zlatko Brievac]

Hello Frederik,

thanks for your reply, but now I have a problem with logging in. I will try it tomorrow. I was expecting a quicker solution.

[Mark Carson]

Hello Zlatko,

Here is an example that should help.

 

Here is an example using SQRTV10 as a predictor. If you use the Predict Dependent Variable functionality and the values from the first case of the input data, you get the "Predicting Values for (HeartDisease)" spreadsheet at the bottom left, which shows how the regression coefficients (b) values map to the equation using the input values and how to transform the input value for the SQRT row before applying it in the equation. Then it shows how you mutliply the b * the values, add them together with the intercept, and get the predicted value (produced by this Summary Predictions spreadsheet if you click Ok on the dialog in the screenshot, which takes you to this screen:

 

Hope this helps.

Mark

[Zlatko Brievac]

Thanks, Mark for your effort. I also have a spreadsheet for my case, but I don't know how to write a mathematical equation based on this spreadsheet. Could you please write a mathematical equation for the example you showed me in the last email? I know how to write part of the equation:

Sylstolic blood pressure = 142.8235 + 0.8504 * Tobacco Intake + 5.73 * LOL cholesterol + ?

So I don't know how to write the last part. I hope it's more evident to you what is bothering me.

Kind regards,

Zlatko

[Zlatko Brievac]

?

[Zlatko Brievac]

Could you please write a mathematical equation for the example you showed me in the email from 4 days ago? Is it possible to get help or not?

[Mark Carson]

Hello Zlatko,

To finish what you are looking for, it is: Sylstolic blood pressure = 142.8235 + 0.8504 * Tobacco Intake + 5.73 * LOL cholesterol - 1.86195 * (SquareRoot(StressTypeABehavior))

StressType-A Behavior was V10 in the original dataset, so the Variable SQRV10 in the bottom left spreadsheet refers to the square root of that variable.

Hope this helps.

Mark

[Zlatko Brievac]

Hello Mark.

According to the formula you completed, the predicted value would be 153.422865

So it would not be 147.189 as in the picture...?

That is what worries me!

Zlatko

[Zlatko Brievac]

Sorry Mark your formula is fine.

Thank you very much.

Zlatko

[Zlatko Brievac]

Hello Mark, thanks again for your replies and for your patience with me.

Kind regards,

Zlatko

[Mark Carson]

You are welcome Zlatko. The QA manager figured this out and sent the information. I will send your thanks along to the QA manager.

Have a great weekend!