Questions tagged [statistical-inference]
The area of statistics that focuses on taking information from samples of a population, in order to derive information on the entire population.
3,774
questions
-1
votes
0
answers
22
views
In Bias-variance decomposition of Mean Squared Error, why must one assume the error $\epsilon$ has $\text{Var}(\epsilon) = \sigma^{2}$?
I'm trying to understand bias-variance decomposition. Proofs of the decomposition often start by stating that the data conforms to the statistical model $y = f(x) + \epsilon $, where $\epsilon$ is an ...
0
votes
0
answers
16
views
Maximum Likelihood of a uniform distribution centered at a random point
I have to compute the maximum likelihood estimator for a uniform distribution centered at a random point.
I have a random variable $A \sim \mathcal{N}(\mu,\sigma^2) $ and a random variable $B \sim Exp(...
0
votes
0
answers
43
views
Central limit theorem application to general functions
Let $X_n$ be a $\mathbb{N}$-valued random value. Suppose we have that
\begin{equation}
\frac{X_n-g(n)}{\sqrt{g(n)}}\sim N(0,1)
\end{equation}
for a function $g(n)$. This looks a lot like the Central ...
-1
votes
0
answers
21
views
Construct a confidence interval for $\theta$
Let $X_{1}, \cdots, X_{n}$ be a random c.i.i.d sample such as, given $\theta$, $X_{1} \sim \mathcal{N}(0,\theta)$. Construct a confidence interval for $\theta$ using asymptotic results.
This question ...
0
votes
0
answers
18
views
What's a good predictive modeling approach for a very small dataset (50-100 rows) [closed]
I want to model the performance of specific athletes, but instead of taking the big data approach and letting the model generalize (something I've already done), I want to create a high bias model ...
0
votes
0
answers
32
views
Estimating integral $\int f(x)pdf(x)^{\alpha} dx$ involving some power of a probability distribution
Suppose a random variable $X$ has an unknown probability distribution $p(x)$ which we can draw samples $X_1,\ldots,X_n$ from. For a known function $f$, under some sufficiently nice conditions, we may ...
0
votes
0
answers
32
views
Show that $T = \max_{i} (X_{i}/i)$ is a sufficient statistic. [closed]
Let $X_{1}, \cdots, X_{n}$ c.i.i.d such as, given $\theta$, have the following p.d.f:
$$
f_{X_{i}}(x|\theta) =
\begin{cases}
e^{i\theta - x}, & x \geq i\theta\\
0, & x < i\theta
\end{...
0
votes
0
answers
40
views
Physical interpretations of covariance and correlation
From what I know, theoretically, covariance is a measure of the degree to which two variables change together. A positive covariance indicates that the variables increase or decrease together, while a ...
2
votes
0
answers
33
views
Non-existence of Lebesgue probability densities
A standard result taught in mathematical statistics courses is that the multivariate gaussian only has a density if the covariance matrix is non singular: i.e. if $X \sim N_p(\mu, \Sigma)$ and $\Sigma$...
1
vote
0
answers
30
views
What does "spectral representation of a Gaussian Process" mean?
In a class about statistical learning, the professor was talking about various theorems like Mercer's theorem, Bochner's theorem, and representer theorem in context of Gaussian Processes (GPs). Two ...
0
votes
1
answer
64
views
Maximum Likelihood for $\beta_1$ Using Partial Derivatives
Assume $y_i = \beta_0 + \beta_1 x_i + e_i$ and $e_i\sim N(0,\sigma^2)$. One way to estimate $\mathbf{\beta}$ is via maximum likelihood estimate. I saw from elsewhere that $\mathbf{\beta}$ could also ...
2
votes
2
answers
64
views
Showing sufficiency of a statistic (the hard way)
Given $X_1,\dots,X_n$ i.i.d. from $P_\theta$, the usual way of showing that a statistic $T$ is sufficient is to use the factorization lemma as opposed to computing the distribution of $(X_1,\dots,X_n)|...
0
votes
1
answer
26
views
conditional distribution of sample given maximum
Given an i.i.d. sample $X_1,\dots, X_n$ from the uniform distribution on $[0,\theta]$, and denoting their order statistics by $X_{(1)} < X_{(2)} < \cdots < X_{(n)}$, it is easy to show that $...
-1
votes
1
answer
13
views
Solved the problem using the echolan form. However, I am currently struggling to find the variables associated with this solution.
The matrix is
$$\left(
\begin{array}{cccc|c}
4 & -1 & 1 & -1 & 3 \\
1 & 2 & -3 & 2 & 2 \\
2 & -5 & 7 & -5 & -1 \\
7 & -4 & 5 & -4 &...
0
votes
1
answer
38
views
How can I get the average value of a class A∩B knowing the average values of classes A and B?
I am trying to model the population density for my metropolitan region, but the published census data does not meet the precision requirements I need. The most accurate data I have avaiable are ...
1
vote
0
answers
31
views
George Casella book on Statistical Inference ; question 4.28
In question 4.28, it says let $X$ and $Y$ be independent standard normal random variables. Find the distribution of $X/|Y|$. I can understand the answer is Cauchy(0,1) distribution. Then question (c) ...
0
votes
0
answers
11
views
Confidence interval for prediction
If we have the following linear model for the response $y_i$, how does one compute a confidance intervall for the expected value of $y_i$?
1
vote
0
answers
41
views
Solutions to Rao's book [closed]
I hope this is the appropriate forum and the questions hasnt been asked already but does anyone know if there is a solution manual for Rao's 'Linear statistical inference and its applications'(2nd ed)?...
3
votes
2
answers
37
views
I have trouble solving the following statistics exercise:
A company produces lightbulbs with an average lifetime of 1000 hours and standard deviation of 50 hours. Find the probability that in a sample of 100 bulbs there are at least two which stop working ...
1
vote
0
answers
15
views
Is statistical inference just probability applied to the results from descriptive statistics?
I have that doubt due to I've been told that this branch of statistics infers from the data, understanding inferring as theestimate of a result, so I guess you use probability to make the estimates. ...
1
vote
0
answers
19
views
Conditional probabilities and conditional validity
In the paper "Conditional validity of inductive conformal predictors" by V. Vovk, the author considers a condition of the form $$P_{X\times Y}(Y\in\Gamma(X)\mid X=x)\geq1-\alpha,$$ where $\...
0
votes
0
answers
45
views
Expectation of a MM estimator [duplicate]
So I want to calculate the expectation of my estimator to see if it is biased or not.
Data:
$\bar{X} = \frac{1}{n}\sum_{i}x_i$
$E[\bar{X}]= \frac{\theta}{\theta + 1}$
$\hat{\theta}_{mm}$ = $\frac{\bar{...
0
votes
0
answers
16
views
Find a sufficient statistic for a non parametric model.
So I'm trying to solve this question but I have no clue about how to go about it, since I haven't done any non parametric sufficient statistics.
Let $X_1, \ldots, X_n$ be an i.i.d. sample, where each $...
0
votes
0
answers
26
views
How to do Hypothesis testing?
I have two time series data.
Electric vehicle demand at a station (whole number time series) every hour called "Clstr 175"
Weather description (categorical time series) at the station every ...
0
votes
0
answers
11
views
DIC, WAIC: non-invariant posterior mean, "negative effective dimension"
I am going through some lecture notes that unfortunatly do not contain any examples.
"However, the DIC relies on the posterior mean as a point estimate, is not invariant
with respect to ...
0
votes
0
answers
12
views
conjugate normal prior for normal distribution: sigma known vs sigma unkown - how to derive posterior
Let $X_1, ..., X_n$ be independent, normally distributed random variables with unknown mean μ and
known variance $\sigma^2$. Further, assume the mean has the following prior distribution
μ ∼ $N (m, s^...
2
votes
2
answers
50
views
Algebraic manipulation question coming from Method of Moments application
So today in my statistical inference class the professor wrote on the board:
Using the Method of Moments:
$$\sigma^2 = E[Y_1^2|\theta] - E[Y_1|\theta]^2$$
$$= \frac{1}{n}\sum_{j = 1}^{n}{Y_j^2} - \...
-1
votes
1
answer
27
views
Building Confidence Interval for Population Proportion [closed]
I'm having trouble understanding how the 95% confidence interval equation for population proportions gets simplified.
Here is the initial equation:
$$
Pr(\overline{X} - 2\hat{SE}(\overline{X}) ≤ p ≤ \...
0
votes
0
answers
18
views
Can (1 - p-value) be the weight of a graph's edges?
I understand that the p-value can be translated as the strength of the statistical tests' results.
From this perspective, can (1 - p-value) be the weight of an undirected graph's edges (like an ...
0
votes
0
answers
17
views
How does the [UK] Office of National Statistics make allowances for incomplete census information?
I entered into a rather ill-tempered argument with a stranger yesterday. It started going downhill when he asserted that there were in fact 90 million people in the United Kingdom, not the 60 that I ...
1
vote
3
answers
108
views
Is $(X_1+X_2+...+X_n)/n$ a random variable?
If $X_1,X_2,...,X_n$ are i.i.d random variables and are all discrete/continuous, then is $(X_1+X_2+...X_n)/n$ also a random variable?
My attempt: For continuous type, I guess it is a random variable. ...
0
votes
0
answers
18
views
How to determine sample size for Generalized Linear Mixed Models?
I want to design a cross-over study (clinical) to assess the performance of different treatments.
The response of the study is binary outcome: success or failure.
I have two fixed effects and one ...
1
vote
1
answer
49
views
Questions about significance level and Type II error
I am learning tests of statistical hypotheses by myself. The book I used is Probability and Statistical Inference by Robert V. Hogg, etc.
I want to confirm some of my understandings are right or not.
$...
2
votes
1
answer
35
views
Coincise introduction to background for semiparametric statistics
I plan to study the theory behind Targeted Maximum Likelihood Estimation, Doubly Robust Estimation, and Semiparametric Theory. I have a background in bioinformatics: I took courses in basic linear ...
2
votes
2
answers
74
views
Definition about probability mass function by Rober V. Hogg
The following definition of pmf is on page51 from Probability and statistical inference by Robert V. Hogg, etc.
The pmf $f(x)$ of a discrete random variable X is a function that satisfies the ...
1
vote
1
answer
47
views
Can Strong LLN and Weak LLN apply to continuous distributions?
Can Strong LLN and Weak LLN apply to continuous distributions? Or it can only apply to discrete distributions?
Representation of LLN: $X_1,X_2,\ldots,X_n$ are i.i.d, and their expectation values are ...
0
votes
1
answer
37
views
Asymptotic distribution of sample variance
Asymptotic distribution of sample variance $S^2$ for an exponential distribution $\mathcal{E}(\frac{1}{\theta})$
$S^2= \frac{1}{n}\sum (X_i-\bar{X})^2$
$\mathbb{E}(S^2)= \frac{n-1}{n}\theta^2$
$\sqrt{...
1
vote
1
answer
36
views
Regression relation to casual relationship
If the correlation coefficient of two variables is 0, can there still be a causal effect between them? And can the causal relationship between these two variables be studied by regression analysis?
0
votes
2
answers
60
views
Analytic form of an ROC curve
I'm studying the problem of combining two sensors for anomaly detection. I want to analyze its performance by the ROC curve. Now I have obtained a parametric equation about the ROC curve:
$$(x,y) = (...
-1
votes
1
answer
67
views
Why is the variance used so often in statistics instead of other higher moments? [closed]
I am learning statistics recently. By far, many statistical tests I saw, e.g. F-test, ANOVA, uses variance as their components. Hardly can I find any statistical test that uses third moment or even ...
0
votes
0
answers
12
views
How we get the basis vector of the following statistical manifold?
Let me define it first,
Let $S=\{p(x;\xi);\xi\in E\subseteq \mathbb{R^n}\}$ be a statistical manifold. We have the basis vector as $\partial_{i}=\frac{\partial}{\partial\xi^{i}}|_{i=1}^{n}$.
Now let ...
1
vote
0
answers
45
views
Find $\pmb{\widehat{\beta}_1}$ and $\pmb{\widehat{\beta}_2}$, two different solutions to the normal equations $\,\mathbf{X'X\pmb{\beta}= X'Y}$.
This is my attempt at a solution.
Please help complete the solution, and I will be grateful to everyone who helps me
my try.
0
votes
0
answers
14
views
Alternative estimators for variance question
Question
Consider another estimator of $\sigma^2$, based on the inter-quartile range of the sample. Let $Y_{i}$ be the $i^{th}$ smallest observation in the sample, or the $i^{th}$ order statistic. ...
0
votes
0
answers
17
views
estimator for exponential distribution?
Let $(X_1,\cdots,X_n)$ be a sample iid of law that belongs to the family of law $\varepsilon(\lambda)$ the question is to compare between two estimators.
the first is $\frac{1}{n}\sum X_i$
they found ...
2
votes
2
answers
40
views
Questions about the relation between convergence in distribution and convergence in probability
I have two sequences of random variables $\{ X_n\}$ and $\{Y_n \}$. I know that
$X_n \to^d D, Y_n \to^d D$. Can I conclude that $X_n - Y_n \to^p 0$?
If I cannot, what other conditions do I need for ...
1
vote
0
answers
21
views
Fisher information matrix of the binomial distribution
The definition of fisher matrix from the book "Methods of information geometry by shun-Ichi-Amari" is as follows
Let $S=\{p_{\xi}\}$ be an n-dimensional statistical model. Where $\xi$ is a ...
-2
votes
2
answers
55
views
Distribution of a sum of normal random variables [closed]
Let $X_1, X_2, ..., X_n$ be independently distributed variables where $X_k \sim N(k\mu, 1)$ for $k = 1, 2, ..., n$ and $\mu\in\mathbb{R}$ unknown.
I calculated that the ML estimator for $\mu$ is equal ...
0
votes
0
answers
18
views
MLE as a pivot for a sample of normally distributed random variables
Let $X_1, X_2, ..., X_n$ be independently distributed variables where $X_k \sim \mathcal{N}(k\mu, 1)$ for $k = 1, 2, ..., n$ and $\mu\in\mathbb{R}$ unknown.
I calculated that the MLE for $\mu$ is ...
0
votes
0
answers
13
views
Why do we bother converting test statistics from st. normal distributions to chi-squared?
Examples are the wald estimator which we obtain by converting a standard normal distributed random variable to a $\chi^2$ distributed variable by taking the square. I am wondering why we bother with ...
0
votes
1
answer
37
views
Soft: Understanding the difference between machine/statistical learning and parameter estimation
Say we have a known model $M$ with unknown parameters and more specifically, $M$ is a parametric model.
Parameter estimation on $M$ is applying an appropriate method for estimating the parameters.
My ...