Creating samples from a well-specified population using a probabilistic method and/or producing random numbers from a specified distribution.

learn more… | top users | synonyms

0
votes
1answer
16 views

How to choose the training, cross-validation, and test set sizes for small sample-size data?

Assume I have a small sample size, e.g. N=100, and two classes. How should I choose the training, cross-validation, and test set sizes for machine learning? I would intuitively pick Training set ...
1
vote
0answers
28 views

Has anyone publicly shared an implementation of RUSBoost in R?

There's no package available on CRAN, so I was hoping someone in the community had written their own function/package. I see it's been done in MATLAB, so I may just have to start with that and write ...
0
votes
0answers
9 views

How to calculate survey sample with more than one outcome variables

I'm working on a proposal and that needs survey design. The basic question is like this: The study seeks to examine factors (e.g., income) associated with the use of health checkup services of workers ...
0
votes
0answers
12 views

How to handle large .csv file in R? [migrated]

I have a large(>100,000) single column floating point time-series data. I want to find structural changes within the data with respect to time( in my case index). In-order to do that, I am using R ...
-1
votes
2answers
34 views

How do you draw a random sample of unique IDs in a large dataset?

Rookie here -- I have a a large data set with about 75,000 observations, and 2000 unique IDs. Therefore, each unique ID has about 37 observations. I'm trying to draw a random sample of unique IDs, say ...
0
votes
1answer
38 views
+100

Is there an equivalent to LHS for a discrete input space?

I was wondering if there is a method that is equivalent to Latin Hypercube Sampling when the input space you are trying to sample from are a finite discrete set of possible values. For example, if I ...
4
votes
2answers
58 views

Do optimization techniques map to sampling techniques?

From any generic sampling algorithm, one can derive an optimization algorithm. Indeed, to maximize an arbitrary function $f: \textbf{x} \rightarrow f(\textbf{x})$, it suffices to draw samples from $g ...
0
votes
0answers
6 views

Missing values on clinical scoring system in SPSS

I have a SPSS database with 747 patients with sickness symptoms and lab results as variables. With those symptoms and results I have generate two scoring systems. I sum the scores of all the variables ...
1
vote
0answers
21 views

ANOVA's and reaction time data

I've been reading a lot on the effect of perceptual load on selective attention but the analysis confuses me somewhat. Some of these studies flag 20-40ms differences between conditions as significant ...
0
votes
0answers
17 views

sampling plan for various number records

I have a set of data. For example, 10,000 observations. Each set has the same format. Each set has 3 to 10,000 records. The question is, what kind of sampling method is reasonable to do with this set ...
0
votes
0answers
7 views

unbiased sampling of expecation over maximization operator

My problem setting is as follows, I have a set of M random variable X = \{ {X_1},{X_2},...{X_M}\} where each variable X_i is estimated via stochastic sequence ...
2
votes
1answer
30 views

How to interpret if my sample statistic is way out in the tail of the bootstrap distribution

I use bootstrapping to generate the distribution / histogram of my sample statistic and find out that the value of my real sample statistic is way out in the tail. What does this mean? Does it mean ...
0
votes
1answer
28 views

Multiple measurements of a single datapoint: what to do with variability?

I am interested in a general question about research design, but I will explain it on a specific example. Suppose I want to determine if one group of people has higher heart rate than the other ...
0
votes
0answers
7 views

sample generative model from a chain/tree

I have a tree of states and I would like to sample from this tree based on pure birth process; however, I don't know how exactly I can do this; so far I have done this; I simplified my problem; the ...
3
votes
0answers
28 views

Sampling methods and parallelization

A couple of years ago I learned about recent work in parallelizing slice sampling methods. More recently, I have read great things about NUTS and Hamiltonian Monte Carlo methods (HMC) in general (e.g. ...
1
vote
0answers
21 views

How are deterministic/logical nodes represented mathematically in directed graphical models?

Various software for performing inference on graphical models support logical(a.k.a deterministic) nodes. PyMc, Winbugs, Smile are the ones I'm aware of. Based on the different inference methods ...
2
votes
1answer
36 views

Is Slice Sampling a special case of Gibbs Sampling?

I read on this thread the following: If you can use both the gibbs sampler and slice sampling to sample from a posterior I would use the Gibbs sampler as the slice sampler seems unnecessary to ...
1
vote
1answer
17 views

choosing one pair randomly from repeated pairs

I have two vectors, in my case all states in the US, and another vector, which indicates the states that it borders. States in vector 1 repeat for each state it borders. For instance, there would be 3 ...
2
votes
1answer
42 views

Rejection-Sampling of Exponential Distribution

Consider the following question. Consider the generation of random numbers following an exponential distribution with some known mean. Give three reasons why the rejection-acceptance method would ...
0
votes
1answer
30 views

Does Normal distribution theory originate in the psychometric literature, especially reliability theory?

The basic statistical literature does not talk about the exact background of the normal distribution. Is the basis of this assumption in psychometry or it has an origin in pure statistics i.e. ...
1
vote
1answer
111 views

Gibbs Sampling Detecting Change point in time series

I was reading through this one page paper on using Gibbs sampling for detecting a change point in a time series like data. While I understand the part where the $\lambda$ and $\phi$ are chosen from a ...
3
votes
4answers
270 views

How exactly does null hypothesis testing work

I'm trying to gain a better understanding of how null hypothesis testing works. I have 3 questions related to the code below: Am I right in saying the probability of each t value in ...
1
vote
1answer
56 views

Sampling from a lognormal distribution

Suppose we are given $\mu$ and $\sigma$ for a lognormal distribution with random variable $X$. $\mu$ is the mean of the variable's logarithm and $\sigma$ is the standard deviation of the variable's ...
0
votes
0answers
19 views

Finding the fewest needed samples for a regression method

I have a regression method that I have applied with success to tens of thousands of well-observed objects. I'm looking to estimate how well it works on poorly observed objects. I selected 50 objects ...
1
vote
0answers
67 views

How to Improve the Design of a Survey of Multiple Rare Variables Using Rare Variable Location Information

I am attempting to update the design of an annual survey which measures the percent-deficient of 26 different asset inventories. Most of these assets don't occur on every sampling unit and some of ...
0
votes
0answers
19 views

Learning theory for search data?

Has learning theory ever been applied in practice for search log data? If so, what are some findings about generalization/learnability from this data? I'm interested in generalization about an ...
3
votes
1answer
27 views

How to sample from the distribution of a Gaussian scale parameter

I would like to be able to sample the standard deviation of a multidimensional Gaussian distribution of dimension $n$; that is, given some $\phi$, I would like to sample $P(\sigma | \phi) \propto ...
2
votes
0answers
34 views

Gibbs sample from AR(1) of exogenous input

I am trying to fit a model where there is a sequence of exogenous "shocks", $X_1, X_2, ..., X_T$, and a AR(1) of these shocks explain $Y_1, Y_2, ..., Y_T$. Specifically, Data (known): $X_1, X_2, ...
3
votes
1answer
55 views

Do you need to report the standard error of the mean when the sample is the population?

The other day, my tutor told me that whenever a mean value is reported, the standard error of the mean must be reported alongside. But surely this isn't always the case? For example, let's say we ...
2
votes
2answers
80 views

Why is it desirable to have low auto-correlation in MCMC?

I keep reading about the need to check for autocorrelation in MCMC. Why is it important that the autocorrelation is low? What does it measure in the context of MCMC?
1
vote
0answers
27 views

Using bootstrapping simulation to compare estimators in survey

I would like to estimate the mean of a population and select a best estimator with minimum variance of the estimated mean. Suppose that I have two estimators est1 and est2, and they could refer to any ...
3
votes
2answers
64 views

Brewer's method for sampling with unequal probabilities with n>2

I'm trying to use Brewer's method to sample 12 units out of a population of 73. I read on Brewer and Hanif's "Sampling with unequal probabilities" that the probability must be proportional to ...
1
vote
1answer
34 views

How a quota sample should be executed?

Let's say I have a population stratified by gender and age group. Having their proportions, how should I choose the sampling units for interview? How do I ensure the defined quotas are met? If the ...
0
votes
0answers
20 views

What is the best ensemble sampler for highly correlated parameter space?

I have a likelihood that I want to estimate the free parameters for it and I am using MCMC to estimate the parameters. Two of the free parameters are positions and I defined uniform priors and one has ...
1
vote
0answers
32 views

Simpson's Paradox with new sample

Suppose you constructing model whose training data is cumulative in nature; meaning each year you can add new observations with all prior observations being kept the same. (e.g. training set is ...
0
votes
0answers
32 views

How many samples do I need to prove that a classification algorithm is better than another?

I have two algorithms A and B, used to automatically classify each of N elements into K categories, N and K both being in the millions. Neither A or B is perfect, but it is relatively easy for a human ...
0
votes
1answer
33 views

Leave-one-subject-out cross validation in Caret

Hi Dear Colleagues, I wonder how to correctly setup a leave-one-subject-out cross validation (LOSO) for train() function in caret. Here is my example code: ...
0
votes
0answers
27 views

Estimating number of unique people

Assume from a prior experiment with have a known truth table of misclassification of logins on an individual basis. $$ \begin{array}{c|lcr} \text{Truth}/\text{Observed} & \text{Al (M)} & ...
2
votes
0answers
49 views

Birthday “Paradox” — with a different perspective

Background: Many people are familiar with the so-called Birthday "Paradox" that, in a room of 23 people, there is a better than 50/50 chance that two of them will share the same birthday. In its more ...
1
vote
0answers
18 views

What are the statistical implications of assigning treatment and control in time series study under 2 different approaches?

I have a project wherein my group wants to measure the effect of counseling at-risk students on their academic outcomes. There are 1,000 students who will be enrolled during the current academic ...
0
votes
1answer
44 views

What are the null and alternate hypotheses? Will you use a left, right, or two-tailed test?

Ben has a coin which he claims is weighted in a way so that when he flips it, heads appears more often than 50% of the time. He tries to prove it to you by flipping the coin 100 times, which ...
3
votes
1answer
71 views

Optimal Allocation

A campus population of size N=9000 is to be surveyed by a stratified sample for the prevalence of a certain disease based upon three strata of respective sizes $N_h$ = 1000, 3000, and 5000 for h = 1, ...
2
votes
1answer
26 views

Expected coverage of a set of random samples

For a set A, I'm running 8 independent random samples, each with a probability of 1/8=12.5% and is without replacement. I know that the set formed by the union of these 8 samples will be of a size ...
3
votes
1answer
40 views

Pearson Correlation for Clustered data

I have a sample that was drawn using cluster sampling. Let say I have variables x and y, and I would like to check the significance of correlation between those two variables in SPSS 20 taking into ...
0
votes
0answers
17 views

Sampling Distribution of mean for Poisson Distributions

Suppose I have random variables $X_i$ which are Poisson distributed with mean $\mu$. I m interested in the sampling distribution of the variable $\frac{X_1+...+X_n}{n}$. We now that as $n$ goes to ...
1
vote
1answer
31 views

Verifying representativeness of a sample, after simple random sampling

I just used the standard formula to determine the sample size of a sample to match the mean of a pupulation with an error margin of 3 percentage points and with a 90% probability. I know would like ...
0
votes
0answers
15 views

How to improve results when using sampling in skewed binary classification?

I am using a data set with 18 features with True/False output (Related to mobile ad targeting). True values occurs only 0.4 % of the time. So, I have used sampling to keep the ratio of True and False ...
1
vote
1answer
38 views

sample frame representing the true population

(i) Is it true to say that if the sample frame is not a random selection (such as a convenience sample), then the sample frame is not representative of the true population? (ii) We can always do a ...
0
votes
0answers
19 views

Down-sampling with building models (specifically random forests)

I was wondering if anyone had ever used down-sampling to build random forests with data that has unbalanced classes. Basically down-sampling samples (with replacement) x*min from the population where ...
0
votes
1answer
23 views

Incomplete questionnaires

Is it recommented to delete the cases from a database if the responses are highly incomplete? Is there a percentage that can be taken into consideration? (for example, deleting a case if more than 50% ...