$How to use bootstrap techniques to criticise a linear model?$

Question

$How to use bootstrap techniques to criticise a linear model?$

I have two sets of data to which I have fitted a simple linear model to. How would I use bootstrap techniques using both methods for bootstrapping a regression model (i.e. bootstrap based on residuals & pair bootstrap) to criticise this linear model?

Statistics

Throwaway3

Report

Mathe

0

You definitely want to increase the bounty for this question.
- Daniel90
  
  0
  
  I second that!
- Throwaway3
  
  0
  
  Ahh. I’m bad at examining how much I should offer for a question. I have used R to generate bootstrap statistics - I’m just not sure how I’m supposed to use original/bias/standard error, or if there is anything else I am supposed to be doing, to criticise the linear model. If you (or anyone else) can offer a very simple explanation or even a link to a book/website which I can read to understand what I’m supposed to be doing, will be an acceptable answer for this question.
- Savionf
  
  0
  
  To decide what a good offer is, think about how much time one may need to spend to write a good solution to your problem, and what a fair hourly rate for such skilled individual is. That would give you a pretty good idea.

Answer

The answer is accepted.

Join Matchmaticians Affiliate Marketing Program to earn up to 50% commission on every question your affiliated users ask or answer.

Mathe

0

You definitely want to increase the bounty for this question.

Daniel90

0

I second that!

Throwaway3

0

Ahh. I’m bad at examining how much I should offer for a question. I have used R to generate bootstrap statistics - I’m just not sure how I’m supposed to use original/bias/standard error, or if there is anything else I am supposed to be doing, to criticise the linear model. If you (or anyone else) can offer a very simple explanation or even a link to a book/website which I can read to understand what I’m supposed to be doing, will be an acceptable answer for this question.

Savionf

0

To decide what a good offer is, think about how much time one may need to spend to write a good solution to your problem, and what a fair hourly rate for such skilled individual is. That would give you a pretty good idea.
Throwaway3

0

Ahh. I’m bad at examining how much I should offer for a question. I have used R to generate bootstrap statistics - I’m just not sure how I’m supposed to use original/bias/standard error, or if there is anything else I am supposed to be doing, to criticise the linear model. If you (or anyone else) can offer a very simple explanation or even a link to a book/website which I can read to understand what I’m supposed to be doing, will be an acceptable answer for this question.
Savionf

0

To decide what a good offer is, think about how much time one may need to spend to write a good solution to your problem, and what a fair hourly rate for such skilled individual is. That would give you a pretty good idea.

Answer 1

Answers can be viewed only if

The questioner was satisfied and accepted the answer, or
The answer was disputed, but the judge evaluated it as 100% correct.

View the answer

See if this helps:

What is Bootstrapping?

Bootstrapping is a statistical technique that can be used to assess the uncertainty and variability of a model. As you mentioned in the question, there are two main methods for bootstrapping a linear regression model

The residual bootstrap and
The pair bootstrap

Residual bootstrap
This method involves resampling the residuals of the linear model and adding them to the predicted values to create new samples. These new samples can then be used to fit a new linear model, and the process can be repeated multiple times to create a distribution of models. By comparing the original model to this distribution, you can assess the uncertainty and variability of the model.

Pair bootstrap
This method involves resampling pairs of observations, rather than individual observations, and fitting a linear model to the resampled pairs. This method is particularly useful for assessing the uncertainty and variability of the model when there are strong dependencies between the observations.
By using bootstrap techniques, you can get a better sense of how well the linear model is likely to perform on new data, and identify any potential sources of error or bias in the model.

By comparing the results of the bootstrapped models to the original model, you can assess the robustness and stability of the original model. Additionally, you can use the distribution of the bootstrapped estimates to create confidence intervals for the model parameters, which can be used to make inferences about the population from which the data were sampled.

Kav10

Throwaway3

0

Yes, this is clear and easy for me to understand. Thank you again Kav for helping me with my statistics struggles :)
Kav10

0

Of course! I am glad you liked it.

$How to use bootstrap techniques to criticise a linear model?$

Answer

Related Questions

Search