I have a datasets for 3 months, its about a vehicle movement and the gas consumption. The vehicle should refill gas after 750 km driving. But the data I have is not enough. The first month I have only 137 km, the second month I have 436 km, and the third 300 km. Any idea how to generate more data that have the same patterns in order to reach 750 km more or less.

Thank you in advance.

  • 2
    $\begingroup$ Use more vehicles or wait longer. $\endgroup$ – whuber Apr 30 at 14:27
  • $\begingroup$ Re-sampling, as in bootstrapping, is part of data analysis, does not increase information, that is it cannot augment experimental data. $\endgroup$ – BruceET Apr 30 at 17:36

I would be very careful in generating more data from a small amount of data. The data you generate will be based on assumptions from the small dataset. If the small dataset is too small to be reliable, why would you assume it's big enough to reliable discern the real pattern? You can't artifically increase your power by generating more data like the data you have. If you could, then you could just copy paste your dataset and suddenly you have twice the data with exactly the same pattern.

| cite | improve this answer | |
  • $\begingroup$ Thank you for your response, But I need to increase the amount in order to reach 750 km distance as mentioned above. For sure I cant just copy the data to have bigger data set, I want to know a way to generate data that is expected based on the 3 month data I have. $\endgroup$ – Tamim Abo Saleh May 1 at 12:16
  • $\begingroup$ That is not a valid thing to do. You can't artificially improve your sample size by modeling more. If that was possible, no one would ever collect large data sets, they'd just get a small set and amplify it. You have 2 options: 1) use the data you have, and accept the sample size as a limitation 2) go and collect more data $\endgroup$ – H. Green May 1 at 14:11

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.