The area of statistics that focuses on taking information from samples of a population, in order to derive information on the entire population.

learn more… | top users | synonyms

0
votes
0answers
10 views

Can I use a z-transformation instead of Welch-t-test

I am not a big mathematician and at the moment I am just refreshing some of my statistical knowledge. So, when doing a students t-test with two independent samples, a precondition is, that both ...
0
votes
0answers
19 views

UMVUE using complete and sufficient statistic

Let $X_1,X_2,...,X_n$ be a random sample from a normal distribution with mean $\mu$ and variance $\sigma^2$. I showed that $(\bar X,S^2)$ is jointly sufficient for estimating ($\mu$,$\sigma^2$) where ...
0
votes
0answers
20 views

Maximum Likelihood estimators in linear models

Consider two simple linear models. $y_{1j}=\alpha _1+\beta_{1}x_{1j}+\epsilon_{1j}$ and $y_{2j}=\alpha _2+\beta_{2}x_{2j}+\epsilon_{2j}$ , $ j=1,2,...,n>2$ where $ ...
0
votes
0answers
19 views

What is the problem with this model parameter estimation algorithm?

In a statistical model with parameters $\theta$ and unobserved laten variables $Z$, the model likelihood is $$L(\theta;X)=Pr(X|\theta)=\sum_ZPr(X,Z|\theta)$$ The standard way to estimate $\theta$ ...
0
votes
0answers
27 views

Parameter estimation of Lorenz system (nonlinear dynamical system)

My problem is as follows. I have to estimate parameters of Lorenz system using given data. Lorenz system is described by following system of ODEs: $$ \frac{dx}{dt} = \sigma(x-y) \\ \frac{dy}{dt} = ...
0
votes
0answers
14 views

Show there is no UMP test for $N(\mu,\sigma^2)$

Here's a testing problem. $X_1, \cdots, X_n, indep, \sim N(\mu, \sigma^2)$ where $\mu, \sigma^2$ are unknown. $H_0: \mu \le \mu_0$ vs $H_1: \mu > \mu_0$ where $\mu_0$ is specified. I ...
0
votes
0answers
6 views

Matrix Completion feature vectors

In the noisy matrix completion problem, with enough number of revealed entries say $|E| = nr^2 log(n) $, can we have a bound on the error in the singular vectors of a sub matrix. For, example say ...
0
votes
0answers
23 views

Question about Logistic Regression - 6

I am studying Logistic Regression and I have come across to understating the paragraph below. I kind of can understand, but it makes me confused when I read the sentence in the red circle, "It also ...
0
votes
0answers
15 views

Backwards quadratic programming to infer Q matrix

Consider the standard QP problem: $\arg\min \frac{1}{2}x^TQx +c^Tx$ Say I know the optimal $x$ for a large number of solutions to this problem with various (known) $c$, and identical (unknown) $Q$. ...
0
votes
0answers
16 views

Normal distributions with a 95% confidence interval

A factory produces items with a mean weight of 3kg. A sample of 100 items has mean weight 3.041kg. Assuming the weight of the items is N[a, 0.04] find a 95% confidence interval for a. I understand ...
0
votes
0answers
16 views

Calculating person-time

Suppose that an investigator wants to measure the incidence rate of high blood pressure in the $1997-98$ freshman class of a university. Assume that there were $1000$ entering freshmen and that this ...
0
votes
0answers
30 views

Proving a statistic is sufficient and NOT complete

Let X be a single observation from the distribution , $P(X=x)= \begin{cases} \theta, & \text{if $x$ =-1} \\ (1-\theta)^2\theta^x, & \text{if $x =0,1,2,3,...; 0<\theta<1$} \\ ...
0
votes
0answers
49 views

Sufficient statistic for normal distribution

Let X from a Normal distribution $(\theta,1)$. a) Find a sufficient statistic for $\theta$. b)Is $S_n^2$ sufficient for $\theta$ My answer for part a) The joint p.d.f= $1 \over (2\pi)^{n/2}$$e^{{-1 ...
0
votes
0answers
17 views

Conceptual doubt between observed values and samples

first of all, I know almost nothing about statistics (and I do no like it or I've never found a good reference). Apparently it is usual to write capital letters to denote random variables and small ...
0
votes
0answers
32 views

Which hypothesis test to use

Two identical machines are used to make a special coin. We want to know if they have the same variability. A random sample is taken from each machine : $$ \begin{matrix} MachineA & 135 & ...
0
votes
0answers
23 views

Poisson distribution confidence intervals and hypothesis

I think I have A and B correct but I have troubles with parts C and D. A) What is the p value if we suppose the following : finding golden apples in a tree follows a Poisson P(2) with $\lambda = 2$ ? ...
0
votes
0answers
21 views

Monte carlo formula to compute the approximation of variance of MLE

In the book of "Monte Carlo Statistical Methods", the book gives an approximation formula for the variance of MLE, Later on, the book mentions that this approximation formula can be written as ...
0
votes
0answers
11 views

Is Expectation Propagation (EP) affected by the prior?

I understand EP by reading Minka's thesis: http://research.microsoft.com/en-us/um/people/minka/papers/ep/minka-ep-uai.pdf I'm trying to apply it to solve a Bayesian inference problem. However, I'm ...
0
votes
0answers
21 views

General procedure Vs pooled variance

When we compare means of two group. With independent sample, we have two method, but i don't know when use choose which one and when choose the other. Please help me classify them Two method are: ...
0
votes
0answers
34 views

Nonrejection region- equivalently the area determined by confidence interval for $H_0: \hat \beta=\beta^*$

For $H_0: \hat \beta=\beta^*$ I want to prove that the non-rejection region in level of significance approach Will be eqaul to the area determined by upper and lower bounds in confidence interval ...
0
votes
0answers
17 views

Cross correlation computations

What are useful ways/formula for calculating sample cross correlations (i.e. correlation factors between individual components of two different random variables). Say I have two sample matrices, $X$ ...
0
votes
0answers
18 views

Calculate probability of total population from sample

I have a population of infinity size. Each Unit is Boolean(True/False). I take a random sample of size N of which M elements are True(and N-M are False). I'd like to know what the probability is that ...
0
votes
0answers
25 views

How to find MLE function?

I have two vectors of known values x and y. And the relationship between them is y=sin($\theta$$\cdot$x)+$\epsilon$, $\epsilon$~N(0,1) . The question is how i find the MLE function for $\theta$?
0
votes
0answers
18 views

Stationary point of Unnormalized and Normalized KL-divergence minimization

I have encountered a problem, basically related to the http://arxiv.org/abs/1206.6679 . If I want to minimize the normalized KL-divergence KL(Q||P) with Q a multivariate Gaussian distribution. but in ...
0
votes
0answers
20 views

Jeffery's prior - help

I'm studying Bayesian inference and looking at prior choices. Currently I have looked at Laplace's uniform prior choice and now I am trying to understand Jeffrey's prior. I am having trouble ...
0
votes
0answers
24 views

Multivariate Normal Variables

Sorry if this is a bit hard to read. Not entirely sure how to use the MathJax formatting on the site... The random variables $X_1, X_2, X_3, X_4$ and $X_5$ are jointly multivariate normal. Their ...
0
votes
0answers
18 views

Point estimation for two different group

Let θ be the difference in the mean precipitation from the two groups. Estimate θ. Estimate the standard error of the estimate and produce a 95 percent confidence interval. About mean, I use ...
0
votes
0answers
20 views

Unbiased estimators of $\theta^3$

Let $X_1, ..., X_n$ iid $∼$ Bin($1,\theta$) $0 < \theta < 1$ and $n >= 3$ I'm trying to find two unbiased estimators of $\theta^{3}$, one that is a function of only $X_1, X_2, X_3$ and ...
0
votes
0answers
18 views

Signed rank test help

suppose i have 10 paired observations (which are skipped here). Suppose also that i want to test $H_0$: no difference between these distributions. Against $H_1$ they differ. that is a two tailed test. ...
0
votes
0answers
23 views

Proof of the independence of the sample mean and sample variance using characteristic function

This question is related to the one asked here Joint characteristic function of $\bar{X}$ and $\left\{ X_{1}-\bar{X},X_{2}-\bar{X},\cdots,X_{n}-\bar{X}\right\}$ is ...
0
votes
0answers
17 views

Showing that the sample variance for an SRS is a biased estimator of the population variance?

EDIT: I suspect I may be going about this all wrong, so maybe disregard this. So, I can get this far on my own: $E(\hat{\sigma}^2) = E(\frac{1}{n - 1}\sum_{i = 1}^n (X_i - \mu)^2) = \frac{1}{n - ...
0
votes
0answers
25 views

Computation of conditional probabilities

I am aware this question might not be well formulated, but it is not very clear for me neither, so if anyone could help me explicit it... I observe many examples e_i. For each example, I compute two ...
0
votes
0answers
9 views

Predict values of some numerical vectors by using other numerical vectors with all these vectors in the same vector set

I need to solve a problem about predicting values of some numerical vectors by using other numerical vectors with all these vectors in the same vector set, which is generated by one or more black box ...
0
votes
0answers
67 views

Find MLE of $\alpha$ of $f(x;\alpha)=(1+\alpha x) /2$ (stuck at derivative setup)

$X_1,...,X_n$ is an independent sample with common density: $f(x;\alpha)=(1+\alpha x) /2$ where $-1<x<1$ and $-1<\alpha <1$ I have to find the maximum likelihood estimate of $\alpha$. ...
0
votes
0answers
22 views

Interpretation of Cook's Distance

The Cook's Distance $D_i$ of the data $(Y_i, x_i)$, $i=1,2, \dots, n$ is defined as: $$ D_i := \frac{||\hat{Y}-\hat{Y}_{(-i)}||^2}{p{\hat{\sigma}}^2}$$ where $p$ is the number of parameters, the 'hat' ...
0
votes
0answers
8 views

Statistic (Linear normal model): $X_{hi} \sim N(\alpha_h+ \beta t_{hi}, \sigma^2)$. How to calculate $C_{95}(\beta)$ and $t$-test for $\beta = 2.5$?

Statistic (Linear normal model): $X_{hi} \sim N(\alpha_h+ \beta t_{hi}, \sigma^2)$. How to calculate $C_{95}(\beta)$ and $t$-test for $\beta = 2.5$? I am in the statistical model $X_{hi} \sim ...
0
votes
0answers
18 views

Estimation of a Random Variable , Probability

A machine consist of two engines ,which stop running independent of each other.In one week the probability of the engines stop running is $p_1$ for the first engine and $p_2$ for the second one.So for ...
0
votes
0answers
20 views

Efficient algorithm for point estimation of a dependent random variable

Suppose $X$ is a normal-distributed random variable and $f$ is a known smooth function (possibly quite complicated, with many oscillations). Let $p(y)$ be the pdf of the dependent random variable $Y = ...
0
votes
0answers
17 views

$p$-dimensional confidence set for normal linear model

this is a question about a confidence set in $p$-dimensional space for a normal linear model. If we define the cuboid $C:= \prod_{j=1}^p C_j(\frac{\alpha}{p})$, where $$C_j(\alpha) = [\hat{\beta_j} ...
0
votes
0answers
10 views

Loss specific inference in graphical models

As far as I have seen, in graphical models, the inference (for training or parameter estimation) is done via maximizing likelihood. While in many applications people need loss specific optimization of ...
0
votes
0answers
57 views

Find the MLE of the parameters

I need some help with the following problem: Let $X$ and $Y$ be independent exponential random variables with $$f_X (x| \lambda)={1 \over \lambda} e^{-x/\lambda},x>0,$$ $$f_Y (y| \mu)={1 \over ...
0
votes
0answers
27 views

What fundamental assumption is taken here?

In biology, the following is often used for collecting data from a population: if we have data from $n$ people, over $m$ years, then we have $nm$ years of data for one person (to simplify). To give a ...
0
votes
0answers
9 views

Tukey's half space depth

For multivariate distributions, Tukey's half space median is defined as "the maximizer" of the quantity: inf{P(a'(X-*x*)>=0|||a||=1}, where X~a multivariate distribution F and the infimum is taken ...
0
votes
0answers
25 views

Multiple comparisons Bonferroni

I want to use Bonferroni method to do multiple comparisons, but i would like to know how the statistic is being proved. From what i understood the statistic of Bonferroni is $$ t=\dfrac{(X_i - X_j )} ...
0
votes
0answers
28 views

Hypothesis test on variance of normal sample

This question is a follow-up to this discussion: Distribution of likelihood ratio in a test on the unknown variance of a normal sample Could someone audit the reasoning below? I am trying to derive ...
0
votes
0answers
17 views

Textbooks with worked examples on identifiability and regularity

I'm looking for textbooks that would have worked examples on identifiability and regularity, something similar to a Schaum's outline. Are there any textbooks that have worked examples on these ...
0
votes
0answers
23 views

Bayesian Variable and Model Selection, Books and Review Papers Desired

I'm hoping that the community will be able to suggest some literature for studying this topic. There seems to be very few books on the subject. There are some chapters in some books which provide ...
0
votes
0answers
52 views

Neyman–Pearson lemma for non monotonic spaces

Question: Does the Neyman–Pearson lemma give instructions for how to construct the test when the outcome space is not monotonic? I suspect the answer is NO, but I would like to: Get an affirmative ...
0
votes
0answers
18 views

Logit Nomal Prior Distribution

$$\mu \sim N(\mu_0,\sigma_0)$$ $$ X_i \sim LN(\mu,\sigma_x)$$ Does anyone know any method for finding the posterior distribution $P(\mu|X)$ or at least any idea of how to estimate it numerically. I ...
0
votes
0answers
29 views

Graphical Models, HMM, DGM

I am studying about Graphical Models and I came up with a simple example but I am not sure which kind of technique (HMM, DGM) would be help me with that. Imagine a have three balls that can move ...