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?
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
- 471 views
- Pro Bono
Related Questions
- Weighted average issue
- Secretary problem 2.0
- Finding the probability that a roughly normal distributed will have the highest value among multiple curves
- Statisitical Experimental Design Question
- Help structure linear mixed effects model random effects structure
- Elements in a set
- ARIMA model output
- What is the normal probability distribution function?
Questions at this level should generally come with a bounty.