Analytical sampling 2026-11-05
Analytical sampling selects and collects material from a larger target for chemical measurement. It connects the material actually measured with the batch, location or time period about which a conclusion is wanted. A small, well-measured portion need not represent the larger material merely because its instrument reading is precise.
An original two-region example has90% of a batch's mass at and10% at of an analyte. The batch-average mass fraction isTaking equal masses from the two regions instead produces . That sample answers a different weighted question. Probability sampling or a justified stratified sampling design can connect selection probabilities and weights to the intended target.
A sampling increment is an individual collected portion. Combining increments and reducing material to a laboratory portion can introduce further variation, so the protocol should state where, when, how much and how material was handled. Repeated readings of one prepared solution chiefly test later measurement stages.
For an original nested model, let independent collected portions have sampling variance , and repeated analyses within a portion have variance . Averaging analyses on each of independent portions givesWith stipulated , , and in compatible units, the mean's standard uncertainty is approximately2.02. Increasing laboratory replication alone cannot eliminate the sampling component. Correlated portions or shared biases require an expanded model.
A primary collaborative trial had multiple samplers collect wheat and green coffee from bulk lots. Its contrasting findings for between-sampler and within-condition variation demonstrate why analytical sampling uncertainty needs a design that actually varies the collection stage.[ref]
Collaborative trials of the sampling of two foodstuffs, wheat and green coffee by Michael Thompson and others. Abstract, multiple-sampler design and sampling reproducibility/repeatability findings.
Jerzy Neyman Created 2026-10-31 Updated 2026-11-05
Jerzy Neyman helped formulate statistical inference in terms of procedures with stated sampling guarantees. With Egon Pearson, he developed hypothesis testing that controls type I error while seeking high statistical power, embodied in the Neyman-Pearson lemma. His confidence-interval theory interprets a confidence level through repeated-sampling coverage rather than a posterior probability for a fixed unknown parameter. He also contributed to stratified sampling and the Neyman Type A distribution.
Neyman moved from University College London to Berkeley in 1938. The university's statistics department grew out of the programme he established there and became a department in 1955. Berkeley's departmental history connects that institutional development to his work on testing, confidence intervals and sampling design.[ref]
https://statistics.berkeley.edu/about/history. UC Berkeley Department of Statistics, History, early years 1938–1955
Jerzy Neyman in Warsaw in 1973
. Photograph by Konrad Jacobs, copyright Mathematisches Forschungsinstitut Oberwolfach. CC BY-SA 2.0 Germany, via Wikimedia Commons. Source. Past exam of the mathematics course of the University of Cambridge 2001 iii Paper 23 4 c Solution Created 2026-10-03 Updated 2026-10-14
Let have finite variance and mean . An independent-sample Monte Carlo estimator has variance , so its root-mean-square random error is . Three techniques reduce the variance coefficient rather than changing this ordinary rate.
For antithetic variates, couple with and average the pair. With independent pairs, the variance is , compared with for independent function evaluations. If is monotone in a scalar uniform, the covariance is nonpositive: for independent uniforms ,Opposite fluctuations therefore cancel. The reduction factor at equal evaluation count is , where is the paired correlation coefficient; it can approach zero, but positive covariance would make the method worse. This is appropriate for monotone or otherwise negatively coupled payoffs, with cheap complementary paths.
For control variates, use a jointly simulated with known mean and average . It remains an unbiased estimator for fixed , andCompleting the square gives and minimal variance . For contingent claim simulation, an analytically priced related financial payoff, such as a simpler option or discounted terminal stock, can be a useful control. It works best when correlation has large absolute value and simulation of the control adds little cost. A coefficient fixed from an independent pilot keeps the elementary unbiasedness argument valid; estimating it from the same sample can introduce finite-sample bias.
For stratified sampling, partition the sampling space into strata of probabilities , and independently simulate samples conditional on each stratum. The estimator is unbiased and has varianceWith proportional allocation , ignoring integer rounding, this equals . The law of total variance shows the reduction from ordinary sampling is : fixed representation removes randomness in the stratum counts. At equal per-draw cost, minimizing the displayed variance gives and minimum for positive within-stratum variances. This follows by Cauchy-Schwarz inequality and is useful when a low-dimensional coordinate strongly predicts the financial payoff and conditional simulation is easy.
There is no universal ranking. At comparable cost the variance factors depend respectively on paired covariance, squared control correlation, and within-stratum variation. A nearly perfect control can outperform a weak antithetic pair; an appropriate stratification can be almost exact for a nearly stratum-constant financial payoff. Extra construction and sampling costs must be included in a fair accuracy comparison, and the methods can be combined.
Past exam of the mathematics course of the University of Cambridge 2004 iii Paper 35 4 d Solution Created 2026-10-03 Updated 2026-10-14
Three useful methods are antithetic variates, a control variate, and stratified sampling. Their purpose is to reduce sampling variance at comparable computational cost, rather than merely to take more samples.
For antithetic variates, generate independent pairs and average both payoffs. Since the call integrand is increasing, the two values have nonpositive covariance. Indeed, for an independent copy , twice that covariance is the expectation of , which is nonpositive pointwise. With evaluations, the paired average has varianceFor a control variate, use the same simulated discounted stock value , whose known mean is . Replace the discounted call payoff by , preserving its mean. Minimizing its quadratic variance givesA predetermined coefficient or an independent pilot estimate preserves exact unbiasedness; fitting the coefficient from the same observations requires care.
For stratified sampling, divide into equal intervals, sample independently and uniformly within each, and average their payoffs. If is the conditional variance in interval , this unbiased estimator has variance . The law of total variance gives , so this is at most the variance of ordinary independent samples. Stratification removes randomness in how many samples fall in each region, which is particularly valuable when conditional payoff means vary strongly.
Past exam of the mathematics course of the University of Cambridge 2013 iii Paper 32 2 a Solution Created 2026-10-03 Updated 2026-10-07
Label batches before consulting the random-number table, so their labels are independent of the test results. A uniformly random permutation of , restricted to a smaller label set, induces a uniformly random ordering of that set. This provides a simple rejection-and-restriction scheme:
- For A, label each day's six batches . Use a fresh permutation, discard labels , and test the first remaining batch. Repeat independently on each working day. Each daily batch has inclusion probability , and five batches are tested per week.
- For B, label the fifteen weekly morning batches . Discard label 16 from a fresh permutation and choose the first three remaining labels. Use an independent permutation for the afternoon stratum. Every three-element subset within a stratum is equally likely, and each batch has inclusion probability . Six batches are tested per week.
- For C, label the weekdays . Keep the first label in this range from a fresh permutation and test all six batches on that day. Repeat independently each week. Every day, and hence every batch, has inclusion probability ; six batches are tested per week.
With independent random digits instead, rejection sampling gives the same uniform choices: for A keep only digits ; for C keep only . For B, use uniform two-digit numbers, retain labels , and reject repeats within a three-batch sample. Taking residues modulo six or fifteen from a table whose range is not divisible by that number would give unequal probabilities. These are respectively day-stratified simple random sampling, morning/afternoon stratified sampling, and a one-day design using cluster sampling. The printed word “rest” for A is interpreted as “test”, consistently with the surveillance task. If it instead meant leaving one batch untested, choose that omitted batch uniformly and test the other five; this alternative would test 25 batches per week, with inclusion probability , and has a different testing budget.
Probability sampling 2026-10-07
Probability sampling chooses observational units by a specified random mechanism with known inclusion probabilities. This separates design-based representativeness from the convenience of obtaining particular observations. Simple random sampling, stratified sampling and cluster sampling are different designs with different precision and cost properties.
Sampling design 2026-11-05
A sampling design specifies a probability distribution over the possible samples from a defined population:For sampling distinct labelled units, is a subset of the population. The design includes the selection mechanism, not merely the final sample size. A procedure with repeated draws can instead require a distribution over ordered sequences; the resulting inclusion probabilities concern whether a label appears at least once.
The design determines inclusion probabilities and joint inclusion probabilities throughThese connect the physical sampling procedure to the Horvitz-Thompson estimator and its sampling variance.[ref]
A Generalization of Sampling Without Replacement From a Finite Universe by D. G. Horvitz and D. J. Thompson. 1952, general unequal-probability sampling without replacement
An original comparison uses four fixed values and selects two units. Under simple random sampling, every one of the six pairs has probability . Each , so the Horvitz-Thompson estimator is twice the selected sum. It takes values with probabilities . Its expected value is the true finite population total , and its variance isA second design chooses only or , each with probability . It has the same sample size and the same individual inclusion probabilities, but the estimator is now or , giving variance . Pair selection matters even when individual representation agrees.
A third design chooses uniformly among the four pairs containing one label from each of those two groups. Its total estimate is always for this particular population. This illustrates the potential gain from stratified sampling when groups correspond to relevant variation, not a claim that the third design dominates for every possible population. A reproducible design records the eligible labels and actual random selection rule before interpreting the observed sample.
