Table of random numbers

A simple, time-honoured illustration of the operation of the Monte Carlo approach is one curious way of estimating the constant n. Imagine a circle inscribed inside a square of side a, and use a table of random numbers to determine the cartesian coordinates of many points constrained to lie anywhere at random within the square. The ratio of the number of points that lies inside the circle to the total number of points within the square na l4a = nl4. The more random points have been put in place, the more accurate will be the value thus obtained. Of course, such a procedure would make no sense, since n can be obtained to any desired accuracy by the summation of a mathematical series... i.e., analytically. But once the simulator is faced with a eomplex series of particle movements, analytical methods quickly become impracticable and simulation, with time steps included, is literally the only possible approach. That is how computer simulation began. 

A reaction is simulated by making random selections from the grid. From a table of random numbers, a two-digit random number is selected. If the cell corresponding to this number is occupied by an A, the A is crossed off (it reacts ) and a B is written in the corresponding space in the B grid. If the random number identifies a cell that does not contain an A or that contains a crossed-off A, no reaction occurs and no action is taken. 

TABLET C.dat Section 4.18 Simulated drug content uniformity measurements 10 different means, starting from 46 mg, with two samples of 10 tablets each at every weight. N = 10, M = 20. To be used with HUBER, HISTO, but also CORREL to test for spurious correlations in table of random numbers and with MSD to test for conformance with limits. 

On each sampling, five field samples were taken from each plot by combining soil samples from each of five sampling points. The five points combined in each sample were selected each day from a table of random numbers. Since a new selection of points was made on each day the individual field samples numbered I through V in the Tables do not represent soil taken from the same points on the field. Each field sample can be regarded as composed of soil from a randomly chosen set of sampling points that equally represent the surface of the entire field plot. 

Find a table of random numbers and use it to reorder the experimental design of Equation 15.1. [See, for example, Cochran and Cox (1950).]... 

From the table of random numbers take 20 different sample data with 10 random numbers. Determine the sample mean and sample variance for each sample. Calculate the average of obtained statistics and compare them to population parameters. 

Besides the number of trials as defined by the choice of design of experiments, it is important to determine the number of replicated trials. Replications are necessary to eliminate robust errors and to determine the reproducibility variance or error of experiment. Since the reproducibility variance has in this case been quite reliably determined in previous study, we may therefore accept a minimal number of replications or a single replication (Sy2=1.0). Prior to doing an experiment one should define the sequence of performing trials, which should be random to annul systematic errors or outside effects. By means of a table of random numbers this sequence has been chosen 15 13 10 5 14 4 6 1 7 8 3 2 9 12 11 and 16. The outcomes of these experiments are given in Table 2.231. 

Note that Rowe11 presents a cautionary tale of the dangers of blind usage of the results of least squares correlation of engineering data. He simulated experimental data by selecting values from a table of random numbers and obtained some seemingly reasonable correlations between them. The principal reasons for the apparent reasonableness were... 

It is important that the method of randomization is actually random. Treatment allocation according to day of the week, date of birth, date of admission or by alternate cases, is not random. The investigator will often know what treatment the patient will get if they enter the trial and so these methods are open to bias. Randomization must be based on tables of random numbers or computer-generated random allocation. It is also important that randomization is secure. Central telephone randomization is preferable to other methods, such as sealed envelopes containing the treatment allocation. 

In an analysis of this problem Kaye and Naylor [5] suggest that several containers should be selected in order to obtain a representative sample. These may be selected systematically, i.e. the 100th, 200th, 300th and so on or, preferably, using a table of random numbers. These containers should be examined individually in order to determine whether the variation between containers is at an acceptable level and combined to determine an average. In either case it is necessary to obtain a representative sample from each container. For this purpose Kaye and Naylor recommend a thief sampler but warn that this may easily give a biased sample. 

Both goals require obtaining a random sample. Here, the term random sample does not imply that the samples are chosen in a haphazard manner. Instead, a randomization procedure is applied to obtain such a sample. For example, suppose our sample is to consist of 10 pharmaceutical tablets to be drawn from 1000 tablets off a production line. One way to ensure a random sample is to choose the tablets to be tested from a table of random numbers. These can be conveniently generated from a random number table or from a spreadsheet, as shown in Figure 8-5. Here, we would assign each of the tablets a number from 1 to 1000 and use the sorted r andom numbers in column C of the spreadsheet to pick tablet 37, 71, 171, and so forth for analysis. 

Random Sample—Selection of units chosen from a larger population of such units so that the probability of inclusion of any given unit in the sample is defined. In a simple random sample, each unit has an equal chance of being included. Random samples are usually chosen with the aid of tables of random numbers found in many statistical texts. Reference-Listed Drug [21 CFR 314.3] — Listed drug identified by the FDA as the drug product on which an applicant relies in seeking approval of its abbreviated application. 

After determining the number of observations per day, the analyst must select the actual time needed to record the observations. To obtain a representative sample, observations should be taken at aU times of the day. There are many ways of randomizing the occurrence of the observations. One method is to use tables of random numbers published in many handbooks and textbooks or the simple random number generators programmed in many hand-held scientific calculators. 

In one approach, the analyst may select numbers from the statistical table of random numbers, ranging from 1 to 480. If each number carries a value, in minutes, the numbers selected can then set the time, in minutes, from the beginning of the day to the time for taking the observations. For example, the random numbers 25 and 152 would mean that the analyst should make a series of observations 25 minutes and 152 minutes after the beginning of the shift. If the day begins at 8 a.m., then at 8 25 a.m., an inspection of the work area would begin, followed by an observation made at 10 32 a.m. 

A table of random numbers can be employed to assign a ran order and the trials. 

Of course, in practical applications, the sampling process does not consist of drawing balls from a jar. There are a variety of more elegant procedures, generally based on successive iterations of a predetermined formula. An alternative approach that is useful when the number of samples to be drawn is relatively small is to reference a table of random numbers. Such tables have been developed and the results recorded in tabular format. (A table of three-digit random numbers appears in T le 10.) Inasmuch as the numbers in the table are randomly generated, users may enter the table at any point and proceed in any direction. 

The standard normal distribution results from the special case wherein pi = 0 and area under the curve from —< to + > is exactly 1.0. If one can develop a table of random numbers for a uniform distribution over the interval 0-1, it is possible to map a set of equivalent values for the standard normal distribution, as in Figure 10. The value along the ordinate represents the probability that the random variable X lies in the interval —< to x. For any random number we can compute the equivalent value x. This latter value is called the random normal deviate. 

The results of 10 simulated trials are shown in Table 13. Consider the first trial, for example a random number, 09, is drawn from the table of random numbers in Table 10. As shown in Table 12, this corresponds to the event A = 2500. Next, a new random number, 52, is drawn, which corresponds to the event N = 30. Note that the same random number cannot be used for both random variables because they are independent The third random variable, P, is normally distributed, so a random normal deviate (RND) is drawn from Table 11. This number, 0.464, indicates a simulated value for... 

For example if one divides into one hundred equal intervals corresponding to 0.00 to 0.01, 0.01 to 0.02, 0.02 to 0.03, 0.99 to 1.00 one can associate the set of two-digit numbers from 00 to 99 with each of these intervals of p. Then by some method of chance one picks two digits. Specifically one can choose from a table of random numbers the second two digits of the first entry in the... 

For random sampling, samples are simply taken at random from the whole population of the material. The only requirement of such a random sampling process is that samples are drawn in a way that all parts of the population have the same chance of being sampled. This can be done by means of a table of random numbers or by means of a computer program that produces pseudorandom numbers. If each of the samples taken is analyzed times, the overall variance of the whole procedure, o, is... 

One of the assumptions of one-way (and other) ANOVA calculations is that the uncontrolled variation is truly random. However, in measurements made over a period of time, variation in an uncontrolled factor such as pressure, temperature, deterioration of apparatus, etc., may produce a trend in the results. As a result the errors due to uncontrolled variation are no longer random since the errors in successive measurements are correlated. This can lead to a systematic error in the results. Fortunately this problem is simply overcome by using the technique of randomization. Suppose we wish to compare the effect of a single factor, the concentration of perchloric acid in aqueous solution, at three different levels or treatments (0.1 M, 0.5 M, and 1.0 M) on the fluorescence intensity of quinine (which is widely used as a primary standard in fluorescence spectrometry). Let us suppose that four replicate intensity measurements are made for each treatment, i.e. in each perchloric acid solution. Instead of making the four measurements in 0.1 M acid, followed by the four in 0.5 M acid, then the four in 1 M acid, we make the 12 measurements in a random order, decided by using a table of random numbers. Each treatment is assigned a number for each replication as follows ... 

Step 3 A random number is assigned for each parameter in Step 1. First, random numbers between 0 and 1 are chosen for each variable. The easiest way to generate random numbers is to use the Rand() function in Microsoft s Excel program or a similar spreadsheet. Tables of random numbers are also... 

X1.3.2.5 Samples were coded to mask the presence of duplicates and a table of random numbers ctated the running order of tests. 

---