## just y=ax+b [General Statistics]

Dear El Maestro,

That is changing the logic of statistics Find the right data for your model instead of finding the right model for your data.

For those models, I think R squared is the way to go, although different sample sizes might have an influence depending on the range of samples that you have. How "small" are the samples? Between 30 and 50 observations per sample? Would methods like cross-validation work on the datasets?

What about calculating confidence intervals for the adjusted R squared of each model and use a Forest plot of those CI, one line for each dataset? This might give you an idea of the impact of the different sample sizes in this particular problem and to decide whether just looking at R square would be enough. Depending on the number of datasets, visual inspection and influence statistics might help on deciding which is the best data for the model.

Regards,

David

That is changing the logic of statistics Find the right data for your model instead of finding the right model for your data.

For those models, I think R squared is the way to go, although different sample sizes might have an influence depending on the range of samples that you have. How "small" are the samples? Between 30 and 50 observations per sample? Would methods like cross-validation work on the datasets?

What about calculating confidence intervals for the adjusted R squared of each model and use a Forest plot of those CI, one line for each dataset? This might give you an idea of the impact of the different sample sizes in this particular problem and to decide whether just looking at R square would be enough. Depending on the number of datasets, visual inspection and influence statistics might help on deciding which is the best data for the model.

Regards,

David

### Complete thread:

- Goodness of fits: one model, different datasets ElMaestro 2017-10-06 23:01 [General Statistics]
- Goodness of fits: one model, different datasets nobody 2017-10-07 16:03
- Experimental setup, details ElMaestro 2017-10-07 18:06
- Visualization ElMaestro 2017-10-07 19:07
- multiple regression? Helmut 2017-10-08 17:17
- just y=ax+b ElMaestro 2017-10-08 17:30
- just y=ax+b Helmut 2017-10-08 17:35
- just y=ax+b ElMaestro 2017-10-08 17:50
- just y=ax+b nobody 2017-10-08 20:26
- ANCOVA with R? yjlee168 2017-10-08 21:28
- just y=ax+b DavidManteigas 2017-10-09 10:34
- just y=ax+b nobody 2017-10-09 10:45

- just y=ax+b Helmut 2017-10-10 18:15

- just y=ax+b ElMaestro 2017-10-08 17:50

- just y=ax+b Helmut 2017-10-08 17:35

- just y=ax+b ElMaestro 2017-10-08 17:30

- Experimental setup, details ElMaestro 2017-10-07 18:06

- Goodness of fits: one model, different datasets nobody 2017-10-07 16:03