IID for Parameter Regression
I want to regress some values $v_i$ according to a function $f(x)$.
I know that there is this relationship between $v_i$ and $f(x)$. For every $1 \le i \le n$, there exists a point $(a\cdot i, \space b\cdot v_i)$ near the curve $(x, \space f(x))$ where $a$ and $b$ are constants.
$f(x)$ is a positive, continuously decreasing, smooth function.
$\lim_{x\rightarrow \infty} f(x) = 0$
Is there a proper way to perform this regression and get the constants $a$ and $b$? I have a few ideas, but I am worried that the regression would be incorrect since the distribution of the votes may or may not be iid.
What is the generally correct way to do this regression?
Join Matchmaticians Affiliate Marketing
Program to earn up to a 50% commission on every question that your affiliated users ask or answer.
- unanswered
- 195 views
- Pro Bono
Related Questions
- Proper Test Selection for Repeated Interval Data Collection?
- Relative percentage increase / decrease
- Bivariate Normality questions
- Probabilities/ states question
- Expected Value of the Product of Frequencies for a Triangular Die Rolled 15 Times
- Car accidents and the Poisson distribution
- Classification problem using one or more Simple Linear Perceptron
- Probability maximum value of samples from different distributions
Questions at this level should generally come with a bounty.