2
$\begingroup$

I'm currently doing an internship where I have to calculate incidence ratios for townships for a period of 11 years. I don't have access to the population data for all years and I would like to do some form of interpolation for the in-between years for which I don't have info. I did a simple linear interpolation with approxfun with R. My school advisor told me to use exponential interpolation and gave me a formula that doesn't fit at all. The majority of the townships have small population that doesn't move much.

Here is the population for 3 districts for a 7 seven year period

187  185  182   181  180  179   178
44   43   43    39   42   44    44
2433 2376 2399  2734 2899 2825  3038

Do you know of other ways of interpolating small population data other than linear interpolation which wouldn't require super complicated models?

$\endgroup$
  • $\begingroup$ Define "super complicated models". For example, a Poisson regression model could deal with this easily, but may or may not be something you consider complex. $\endgroup$ – Fomite Oct 1 '15 at 0:31
  • $\begingroup$ I ended up doing an exponential interpolation and estimating the annual growth rate using the growth rate for the 12th year period. Thank you for answering though, I'll look into that for future reference. $\endgroup$ – N F N Nov 9 '15 at 0:32

Your Answer

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

Browse other questions tagged or ask your own question.