Questions tagged [estimators]
A rule for calculating an estimate of a given quantity based on observed data [Wikipedia].
870
questions
2
votes
0
answers
26
views
Model has higher (and closer to 1) $\beta$, but similar $R^2$ and correlation
I have model one which produces prediction $\hat{y_1}$, later I came up with a new model which produces prediction $\hat{y_2}$. I have ground truth $y$. The models are not regression based but they ...
0
votes
0
answers
13
views
Combining information from different quantiles
I have a number of "mostly" Gaussian distributions (in truth a Gauss core and longer tails). I am interested in the width of this distributions.
Given that I do not know the amount of tails ...
2
votes
1
answer
61
views
variance of the estimator of unconditional mean of AR(1) process
AR(1) process is defined as: $y_t=c+\phi y_{t-1}+\varepsilon_t$ where $\varepsilon_t$ is IID with mean zero and variance $\sigma^2<\infty$. For a stationary process, i.e. $\phi\ne 0$, the ...
0
votes
1
answer
48
views
Are all random variables estimators? [duplicate]
My hand-wavey understanding is a random variable is a function from a domain of possible outcomes in a sample space to a measurable space valued in real numbers.
We might denote a random variable from ...
5
votes
3
answers
682
views
confidence intervals for proportions containing a theoretically impossible value (zero)
This is really a hypothetical question not related to an actual issue I have, so this question is just out of curiosity. I'm aware of this other related question What should I do when a confidence ...
2
votes
0
answers
46
views
No Existence of Efficient estimator
I need to prove that given $(X_1,...,X_n)$ from the density $$\frac{1}{\theta}x^{\frac{1}{\theta}-1}1_{(0,1)}$$ no efficient estimator exists for $g(\theta)$=$\frac{1}{{\theta}+1}$.
I have shown that ...
0
votes
0
answers
41
views
Unbiased Estimator of Nugget Effect
Question: I am trying the measure the nugget effect, which is parameterized by $(1-\lambda)$ in the following variance-covariance used to describe the multivariate normal distribution of my n-...
4
votes
1
answer
49
views
How to accurately estimate the probability of a rare event in a large dataset?
I have a dataset of 30,155 names and out of curiosity I verified that the longest name has 68 characters, which is quite big considering the mean and SD were 24.78 and 5.64, respectively. Based on ...
1
vote
0
answers
39
views
Calculating the mean and error for correlated measurements involving different estimators and quantiles
My goal is to find a way to report a mean $\pm$ error for different estimators and quantiles of the same distribution (same measurement).
I am measuring the width of a distribution (Gaussian core and ...
0
votes
1
answer
23
views
Unbiased and consistent estimator with positive sampling variance as n approaches infinity? (Aronow & Miller) [duplicate]
In Aronow & Miller, "Foundations of Agnostic Statistics", the authors write on p105:
[A]lthough unbiased estimators are not necessarily consistent, any unbiased estimator $\widehat{\...
3
votes
2
answers
76
views
Maximum liklihood estimators for simple linear regression with $\sigma^2$ unknown
Suppose that we have the simple linear regression model for the form:
$$Y_i = \beta X_i +\varepsilon_i$$
With the following set of 'classical assumptions' holding:
$E(\varepsilon_i)=0$
$Var(\...
2
votes
1
answer
120
views
Sum of asymptotically independent random variables - Convergence
Let $\theta_N=\frac{1}{N}\sum_{i=1}^N \pi_i\cdot g_i$ where $0<\pi_i<1$ and $0<g_i<1/\pi_i$ such that $\theta_N\overset{N\rightarrow \infty}{\rightarrow}\theta$.
If $X_i\sim Ber(\pi_i)$, I ...
4
votes
2
answers
124
views
Must maximum likelihood method be applied on a simple random sample or on a realisation?
I guess my trouble is not a big one but here it is: when one applies maximum likelihood, he considers the realization $(x_1, \dots, x_n)$ of a simple random sample (SRS), leading to ML Estimates. But ...
5
votes
2
answers
523
views
Asymptotic unbiasedness + asymptotic zero variance = consistency?
Here, Ben shows that an unbiased estimator $\hat\theta$ of a parameter $\theta$ that has an asymptotic variance of zero converges in probability to $\theta$. That is, $\hat\theta$ is a consistent ...
0
votes
0
answers
22
views
Is Coefficient of Variation a valid measure of relative efficiency?
I'm wondering if it is always valid to use Coefficient of Variation (CV) to determine relative efficiency of parameter estimators, and to compute statistically equivalent sample sizes based on that ...
1
vote
1
answer
72
views
Using Rao-Blackwell to improve the estimator of P(X/Y < t)
X and Y are independent N (0, 1) random variables, we want to approximate P (X/Y ≤ t), for a fixed number t.
The first part of the problem was to describe a naive Monte Carlo estimate. I described ...
0
votes
0
answers
18
views
What is the difference between unbiasedness, consistency and efficiency of estimators? How are these interrelated among themselves? [duplicate]
!Efficiency(https://stackoverflow.com/20240427_193105.jpg). Given snapshot of the book states that among the class of consistent estimators, in general, more than one consistent estimator of a ...
6
votes
1
answer
165
views
Terminology clarification about sample moments
According to MathWorld (link): "The sample raw moments are unbiased estimators of the population raw moments".
While in Wikipedia (link) it is said:
...the $k$-th raw moment of a population ...
1
vote
1
answer
34
views
Why can we get better asymptotic global estimators even for IID random variables?
Let $X_1,...,X_N$ be IID random variables sampled from a parametrised distribution $p_\theta$, and suppose my goal is to retrieve $\theta$ from these samples.
We know that the MLE provides an ...
1
vote
0
answers
49
views
Standard practice to show Biased CRBs
I have a problem with four-parameter estimation. I have derived the variances for the estimated parameters using Monte Carlo simulations (numerical ones) and theoretical ones using the inverse of the ...
1
vote
1
answer
54
views
What is the distribution of the unbiased estimator of variance for normally distributed variables?
I must be making some mistake in my derivation of the distribution of the unbiased variance estimator for i.i.d. $X_{i} \sim \mathcal{N}\left(\mu, \sigma^{2}\right)$.
We have $\bar{X} =\frac{1}{n}\sum\...
0
votes
0
answers
18
views
Demonstrating $SU=U(\sigma^2 I+D^2)$ as a Sufficient Condition in Maximum Likelihood Estimation
I am working on an exercise related to maximum likelihood estimation (in the context of principal component analysis) for the distribution
$$p(x) = Gauss(b, WW^T+\sigma^2I)$$
In particular, I want to ...
1
vote
0
answers
23
views
Degrees of freedom for estimation
In the context of estimators, why is it that in general dividing by the degrees of freedom(instead of the sample size) leads to unbiasedness? I see the value in substituting degrees of freedom for ...
0
votes
1
answer
36
views
Assumptions needed for consistency of plug-in estimator
Assume $X,Z$ are random variables and let $x_0$ be a fixed number. I want to estimate $A =\mathbb{E}_{X,Z}[\frac{X}{P(X=x_0|Z)}]$.
If $P(X=x_0|Z=z)$ is known for all $z$ we can apply the LLN and ...
2
votes
0
answers
21
views
When are mean and variance estimates uncorrelated or independent
I know that in the case of the normal distribution, the MLE estimates of the mean and the variance are independent. My impression is that this is a rare property for a distribution to have. Are there ...
5
votes
2
answers
128
views
Sufficient conditions for asymptotic efficiency of MLE
Maximum-likelihood estimators are, according to Wikipedia, asymptotically efficient, that is they achieve the Cramér-Rao bound when sample size tends to infinity. But this seems to require some ...
1
vote
0
answers
58
views
Is there a good review on complete class theorems?
I'm trying to get an overview of the various results called "complete class theorems" and their relatives, especially the ones that say things along the lines of "every admissible ...
1
vote
2
answers
94
views
Covariance of Best Linear Unbiased Estimators and arbitrary LUE
I'm working on a problem involving two linear unbiased estimators $T$ and $T'$ of a parameter $\theta$, defined from a sample $\{X_1, \dots, X_n\}$ with mean $\theta$ and finite variance. I aim to ...
1
vote
0
answers
120
views
Distribution of $F_n^{-1}(3/4)-F_n^{-1}(1/4)$ [closed]
Given $X_1,X_2,...X_n\overset{\text{iid}}{\sim}F$, find the distribution of the sample inter quartile range, $F_n^{-1}(3/4)-F_n^{-1}(1/4)$ in terms of $F$ where, $F_n$ is the emperical distribution ...
4
votes
1
answer
109
views
Probability mass function of sample median (Bootstrap)
Consider a sample $X_1,X_2,...X_n\overset{\text{iid}}{\sim}F$. Let $T_n=F_n^{-1}(1/2)$ be the sample median where, $F^{-1}(x)=\inf\{t:F(t)\ge x\}$ and $F_n(y)=\frac{1}{n}\sum_{i=1}^n\mathbb{I}(X_i\le ...
1
vote
1
answer
69
views
Intuition behind between-group covariance matrix from MANOVA?
Suppose that we have samples from $m$ different $p$-dimensional normal multivariate distributions, where they share a common covariance matrix $\Sigma$ but the mean vectors may be different for each ...
4
votes
1
answer
147
views
Why does not this underlying hypergeometric distribution lead to unbiased estimators?
This example is take from Lippman's "Elements of probability and statistics".
Let N be the number of fish in a lake the warden wants to estimate. He catches 100 fish, tags them and releases ...
4
votes
1
answer
115
views
Verifying mean and covariance estimators of a two-dimensional normal distribution
Here I try to verify estimators of the mean and covariance matrix of the two-dimensional normal distribution $N(\mu, A)$ with $\mu=[-2,3]^T$ and $A=\begin{pmatrix}
5 & 11\\
11 & 25
\end{...
9
votes
1
answer
96
views
Adjusted R2 and bias
Consider the population $R^2$:
\begin{equation}
\rho^2 = 1- \frac{\sigma^{2}_u}{\sigma^{2}_y}
\end{equation}
This equation describes the proportion of the variation in $y$ in the population explained ...
0
votes
0
answers
146
views
What's the difference and relationship between theta, theta star and theta hat?
I understand that $\theta$ is the true distribution parameter (great explanation here). I also know that $\hat\theta$ is an estimator of the true $\theta$ (so for example, MLE is an example of $\hat\...
2
votes
1
answer
77
views
Is an estimator that always have a value of zero is a linear estimator?
Consider a simple linear regression model:
$$Y=\beta_0+\beta_1 X +u$$
Here, we can consider an estimator that does not use any data:
$$\hat{\beta}_1=0$$
That is, regardless of the observed data, the ...
3
votes
0
answers
57
views
What is a measure of hardness-of-approximation by samples?
Suppose there is a large vector $\mathbf{x}$ of real numbers, and I want to estimate a certain aggregate function $f(\mathbf{x})$ by taking a small sample of the population $\mathbf{x}$. I would like ...
0
votes
0
answers
14
views
Optimality criterion for mean estimators
Assume a sample size of $n>5$, a given variance $\sigma^2 > 0$ and a $\delta \in (2e^{-n/4}, 1/2)$.
Proof that there exists a distribution with variance $\sigma^2$ such that for any mean ...
1
vote
1
answer
41
views
For a biased estimator, how does one call the point for which the expected value of the estimator is equal to the observed sample estimate? [closed]
Let $\hat{\theta}$ be a biased estimator whose bias depends on the true value $\theta_0$, such that $E[\hat\theta|\theta_0]= f(\theta_0)\neq \theta_0$. Let $t_{sample}$ be a sample realization of $\...
1
vote
0
answers
42
views
Showing that the estimator of the log posterior used in stochastic gradient MCMC is unbiased
In the SGLD paper as well as in this paper it is claimed (paraphrasing) that
the following estimator:
$$\widetilde{U}(\theta) = -\dfrac{|\mathcal{S}|}{|\widetilde{\mathcal{S}}|} \sum_{{x}\in \...
4
votes
1
answer
124
views
Confidence interval on ratio of estimates for exponential random variables
Given exponential random variable X, the MLE for the scale parameter is $\hat{\beta_x} = \bar{x}$, and the confidence interval for that estimate is:
$$\frac{2n\bar{x}}{\chi^2_{\frac{\alpha}{2},2n}} &...
0
votes
1
answer
82
views
What is the variance decomposition method?
For $i = 1, \ldots, m$ and $j = 1, \ldots , n$ we have observations $x_{ij}$. We can assume that
$$
x_{ij} = y_{i} + z_{ij}, \qquad y_{i} \sim \mathcal{N}(\mu_{y},\sigma_{y}^{2}), \quad z_{ij} \sim \...
2
votes
1
answer
86
views
Maximum Likelihood Estimation for a Unique Probability Density Function
In the context of estimating parameters for a uniquely distributed set of independent and identically distributed random variables, I am examining the following probability density function $ f(x|\...
0
votes
0
answers
21
views
How to show that the influence function of minimum density power divergence estimator with positive tuning parameter is bounded?
In the linked paper, in the influence function section, the term ${u_{\theta}(y)}{f_{\theta}(y)}^\alpha$ is directly called bounded which i do not get the explanation of? Here $\alpha > 0$ is the ...
4
votes
1
answer
96
views
Mathematical Step for consistency
Let me state my problem from the beginning:
Let $i$ be an index representing countries ($i = {1,2,\ldots,N }$), and $t$ represent time, denoted as available data for country $i$ ($t = {1,2,\ldots,T_i }...
0
votes
1
answer
76
views
Unable to estimate AR(p) coefficients and $\sigma^2$
I am currently trying to solve this problem pertaining to the Yule-Walker equations:
Let $\{X_t\}_{t\in Z}$ be a causal autoregressive process given by $$X_t = \varphi X_{t−2} +W_t$$ with $\{W_t\}_{t\...
0
votes
0
answers
27
views
One off the advantages of the bootstrap is that i don't have to worry about having a good estimator?
I'm learning about the theory of estimators and saw that sometimes the analytical formula of the estimator has to be diferent off the formula for the parameter, for example the standart deviation, and ...
1
vote
0
answers
33
views
Large samples property of bayes procedures
I was reading through Wasserman's All of Statistics and I came across this property in the Bayesian statistics chapter:
I think I don't really get what is supposed to be the intuition behind it, and ...
3
votes
1
answer
149
views
Is the sample mean an unbiased estimator of population mean in the presence of autocorrelation?
I've seen previous questions here that the sample mean can be considered an unbiased estimator of the population mean. e.g.1, 2.
While the examples seem to refer to independent sample points, it seems ...
1
vote
1
answer
70
views
How to estimate how heavy a tail is?
Suppose I have data coming from a single variate distribution. I want to estimate how heavy the tail of the distribution is. For example, if the data comes from the Zipf distribution, I would want the ...