Пятница, 28 марта, 2025

Statistical study of variation

A variation of a feature is a difference in the numerical values of a trait in individual units of the population. The size of the variation allows us to judge how homogeneous the group under study is and, consequently, how characteristic the average group is. The study of deviations from the average is of great practical and theoretical importance, since the development of the phenomenon is manifested in deviations.

Statistics are presented in the distribution series. Depending on the feature underlying the grouping of data, attribute and variation series are distinguished. The numerical values of a feature that occur in a given population are called variants of values. Statistical data without any systematization form a primary series.

Example.

CHP No

1

2

3

4

5

Cost price 1 kWh, thousand rubles.

5,8

6,6

5,9

6,7

6,6

If there is a sufficiently large number of variants of the values of the feature, it is necessary to order the primary series to study it, i.e. rank – arrange all variants of the series in an increasing (or decreasing) order.

CHP No

1

2

3

4

5

Cost price 1 kWh, thousand rubles.

5,8

5,9

6,6

6,6

6,7

When looking at the ranked data, you can see that the variants of the values of the feature in individual units are repeated. The number of repetitions of individual variants is called the repetition rate (Statistical study of variation).

By the nature of the variation, discrete and continuous signs are distinguished. Discrete features differ from each other by some discontinuous number.

Table 16 Resource requirements

Distribution of workshop workers by qualification

Tariff

Discharge (Statistical study of variation)

Number of workers Statistical study of variation)

Frequencies (Statistical study of variation)

Statistical study of variation

Accumulated frequencies (Fi)

II

1

0,05

1

III

5

0,25

6

IV

8

0,40

14

V

4

0,20

18

VI

2

0,10

20

Total:

20

1,00

Instead of the absolute number of workers who have a certain rank, you can set the proportion of workers in that rank. Frequencies represented in relative terms are called frequencies and denote Statistical study of variation:

Statistical study of variation.

Frequencies can be expressed in fractions of one or as a percentage. The accumulated frequencies are determined by sequential summation of frequencies.

Continuous features can differ from one another by an arbitrarily small amount and within certain boundaries take any values. To construct a series of distributions of continuous features, the values of the variants are indicated in the intervals “from – to”. When constructing interval series, it is necessary to determine the number of intervals and determine the size of the interval:

Statistical study of variation.

If the variation series is given in unequal intervals, then the absolute and relative densities of the distribution must be calculated to correctly represent the nature of the distribution. Absolute density:

Statistical study of variation,

where Statistical study of variation is the size of the interval.

Relative density:

Statistical study of variation,

where Statistical study of variation is the frequency.

These indicators are used to transform intervals if the data are collected from different populations and processed differently:

Statistical study of variation.

To characterize the size of the variation, special indicators of oscillation are used: the range of variation, the average linear deviation, the mean quadratic deviation, the coefficient of variation.

The range of variation is the magnitude of the difference between the maximum and minimum values of the feature:

Statistical study of variation.

The advantage of this indicator is the simplicity of calculation. The disadvantage is that this indicator relies only on two extreme values of the feature and does not take into account the degree of oscillation of the bulk of the members of the series.

The mean linear deviation is the arithmetic mean of the absolute deviations of the individual values of the feature from the average value.

For the primary series: Statistical study of variation .

For the distribution series: Statistical study of variation .

Since, according to the property of the arithmetic mean, the algebraic sum of the deviations of the individual values of the feature from the arithmetic mean is zero, the absolute values of individual deviations are summed up for the calculation Statistical study of variation , regardless of the sign.

The mean linear deviation shows how much, on average, the individual values of the feature differ from their average value.

The mean square deviation is equal to the square root of the mean square deviations of the individual values of the feature from the arithmetic mean.

For the primary series: Statistical study of variation .

For the distribution series: Statistical study of variation .

The mean linear and mean quadratic deviations show how much the value of the feature fluctuates on average in the units of the studied population: Statistical study of variation>Statistical study of variation. For moderately asymmetric distribution series, the following ratio is established: Statistical study of variationor Statistical study of variation.

Variance has an independent value in statistics and is one of the most important indicators:

For the primary series: Statistical study of variation .

For the variation series: Statistical study of variation .

Hence: Statistical study of variation .

In statistics, it is often necessary to compare the variations of various features. In such cases, an indicator of relative scattering is used – the coefficient of variation:

Statistical study of variationStatistical study of variation.

The coefficient of variation shows by what percentage on average the individual values differ from the arithmetic mean. It is a criterion for the reliability of the average: if it exceeds 40%, then this indicates a large oscillation of the feature and, therefore, the average is not reliable enough.

Linear coefficient of variation: Statistical study of variation .

Oscillation coefficient: Statistical study of variation .

Dispersion has a number of properties.

1. The variance of a constant number is zero. If Statistical study of variation then Statistical study of variation

Statistical study of variationStatistical study of variation.

2. If all variants of one series are increased or decreased by any number, the variance of the new series will not change.

Let Statistical study of variation, but then Statistical study of variation

Statistical study of variation.

3. If all variants of the series are reduced or increased by Statistical study of variation a factor, then the variance of the new series will decrease (or increase) in Statistical study of variation.

Let Statistical study of variation, then Statistical study of variation

Statistical study of variationStatistical study of variation.

The moment of distribution is called the arithmetic mean of certain degrees of deviations of individual values of the feature from a certain initial value. In general, the moment can be written as follows:

Statistical study of variation,

where A is the value from which the deviations are determined;

k is the degree of deviation (order of moment).

Depending on the value, the moments can be calculated of any order, but the moments of the first four orders are practically used.

Any number can be taken as a constant value A. Depending on what is taken as a constant value, the following three types of moments are distinguished:

1) if zero is taken as a constant value, i.e. A = 0, then the moments are called initial. In general, they can be written:

Statistical study of variation and, accordingly, the moments of the first four orders;
Statistical study of variation;
Statistical study of variation – the arithmetic mean of the squares of the variants;
Statistical study of variation;
Statistical study of variation.

2) if the arithmetic mean series is taken as a constant value, i.e. A = Statistical study of variation, then the moments are called central:

Statistical study of variation;
Statistical study of variation according to the property of the arithmetic mean;
Statistical study of variation dispersion;
Statistical study of variation to calculate the excess rate.

3) if any number other than zero is taken as a constant value, then the moment is called conditional:

Statistical study of variation;
Statistical study of variation;
Statistical study of variation;
Statistical study of variation;
Statistical study of variation.

Using the initial moments of the first and second order, it is possible to obtain a formula for calculating the variance:

Statistical study of variation

You can also calculate the variance as follows:

Statistical study of variation

Statistical study of variation

Therefore, the variance can be defined as the difference between the average square of the variants and the square of their mean.

In variation series with equal intervals, the variance can be calculated by the method of moments and by the method of reference from the conditional zero.

The calculation is made according to the formula:

Statistical study of variation,

Where is:

Statistical study of variation– interval width;
Statistical study of variation, x0 is a conditional zero, which is convenient to use the middle of the interval with the highest frequency;
Statistical study of variation – Second-order moment;
Statistical study of variation is the square of the first-order moment.

The units of the phenomena under study can be characterized by such features that some units of the aggregate possess, while others do not. Such a sign is called an alternative one.

The presence of a feature is indicated by one, and its absence by zero. The proportion of units possessing this feature is denoted by p, and the fraction that does not possess it is q. Therefore, p + q = 1, q = 1 – p. The average value of the alternative feature is:

Statistical study of variation.

Thus, the average value of an alternative feature is equal to the value of the fraction of units that possess it.

Let’s determine the variance:

Statistical study of variation

Statistical study of variation.

Example.

Of the 1,000 finished products, 250 were of the highest quality. Define Statistical study of variation.

Statistical study of variation or 25% of the highest quality products.

Statistical study of variation

Statistical study of variation.

To assess the influence of various factors that determine the oscillation of individual values of the trait, it is possible to use the decomposition of variance into components: intergroup and intragroup variance.

The variation due to the influence of the factor underlying the grouping is characterized by intergroup variance, which is a measure of the oscillation of particular averages by groups from the total average:

Statistical study of variation,

where Statistical study of variation are the group averages,

Statistical study of variation – the total average for the whole population,

Statistical study of variation – Number of individual groups.

Variation due to the influence of other factors is characterized in each group by group variance:

Statistical study of variation,

and for the population as a whole, the average of the intragroup variances:

Statistical study of variation.

Therefore, the total variation of the trait in the aggregate should be defined as the sum of the variation of the group averages (due to one isolated factor) and the residual variation (due to other factors). This equality is reflected in the rule of addition of variances Statistical study of variation.

The ratio of intergroup variance Statistical study of variation to the total Statistical study of variation is given by the coefficient of Statistical study of variationdetermination , which characterizes the proportion of variation of the resulting feature due to the variation of the factor feature (which is the basis of the grouping).

The coefficient of empirical correlation characterizes Statistical study of variation the closeness of the relationship between the effective and factor features.

To get an idea of the form of distribution, distribution graphs (polygon and histogram) are built. The number of observations from which the empirical distribution is constructed is usually small and is a sample from the general population under study. With an increase in the number of observations and at the same time a decrease in the size of the interval, the zigzags of the polygon begin to smooth out, and in the limit we come to a smooth curve, which is called the distribution curve.

Statistics investigate different types of distribution. As a rule, they are single-vertex. Polyversity indicates the heterogeneity of the population under study. The appearance of two or more vertices indicates the need to rearrange the data in order to highlight more homogeneous groups.

A symmetric distribution is one in which the frequencies of any two variants equal in both directions from the center of distribution are equal to each other. For symmetric distributions, the arithmetic mean, the mode, and the median are equal. The simplest indicator of asymmetry is based on the ratio of indicators of the center of distribution: the greater the difference between the arithmetic mean and the mode (median), the greater the asymmetry of the series.

Asymmetry index:

Statistical study of variation or Statistical study of variation.

To compare asymmetry in several rows, a relative asymmetry indicator is used.

Statistical study of variation or . Statistical study of variation

Statistical study of variation The value can be positive and negative. If Statistical study of variation, then on the graph such a series will have an elongation to the right (right-sided asymmetry), if Statistical study of variation, then an elongation to the left (left-sided asymmetry).

The steepness characteristic of the distribution is also calculated. This is an indicator of excess. With the same arithmetic mean, the empirical series may be peaked or low-vertex compared to the normal distribution curve. The excess rate reflects this feature:

Statistical study of variation.

If Statistical study of variation the > 0, then the excess is considered positive (the distribution is peaked), if Statistical study of variationthe < 0, then the excess is considered negative (the distribution is low-vertex).

Among the various distribution curves, a special place is occupied by the normal distribution. The normal distribution in a graph is a symmetrical bell curve having a maximum at a point Statistical study of variation. This point is the mode and median. The inflection point of a normal curve is ±Statistical study of variation away from Statistical study of variation. The normal distribution curve is expressed by Laplace’s equation:

Statistical study of variation,

where t is the normalized deviation, . Statistical study of variation

It is established that if the area bounded by the normal distribution curve is taken as 100%, then it is possible to calculate the area enclosed between the curve and any two ordinates. It is established that the area between ordinates drawn at a distance Statistical study of variation on each side of Statistical study of variation, is 0.683 of the total area. This means that 68.3% of all frequencies (units) deviate from Statistical study of variation no more than Statistical study of variationby , i.e. are within Statistical study of variation. the area enclosed between ordinates drawn at a distance of 2Statistical study of variation from Statistical study of variation in both directions is 0.954, i.e. 95.4% of Statistical study of variationall units of the aggregate are within . 99.7% of all units are within . 99.7% of Statistical study of variation all units are within . . This is the rule of three sigmas, characteristic of the normal distribution.

Normal distribution is characteristic of phenomena in biology and engineering. In economics, moderately asymmetric distributions are more common.

When dealing with empirical distributions, it can be assumed that each empirical distribution corresponds to a certain, characteristic theoretical curve. Knowledge of the shape of the theoretical curve can be used in various calculations and forecasts. To do this, you need to determine:

the general nature of the distribution; to construct a theoretical curve from empirical data; determine how close the empirical frequencies are to the theoretical ones.

Let’s introduce the designations:

Statistical study of variation, Statistical study of variation,

where Statistical study of variation = 2.7182 (base of the natural logarithm);

Statistical study of variation = 3,14.

To construct a theoretical curve of normal distribution from empirical data, it is necessary to find theoretical frequencies:

Statistical study of variation,

where Statistical study of variation is a constant;

h is the width of the interval;

Statistical study of variation is the tabulated value that is located by deviations t.

The sequence of calculation of theoretical frequencies is as follows:

the arithmetic mean of the seriesStatistical study of variation is calculated; the mean quadratic deviation Statistical study of variationis calculated; is located Statistical study of variation; according to the found t on the table is Statistical study of variation; calculated Statistical study of variation; each value Statistical study of variation is multiplied by Statistical study of variation.

Among the most important theoretical distributions is the Poisson distribution, which is characteristic of rare phenomena, and with an increase in the value of x, the probability of their occurrence decreases.

The Poisson distribution is as follows:

Statistical study of variation,

where Statistical study of variation .

Then:

Statistical study of variation Statistical study of variation.

Graphically, it looks like this:

Finding the theoretical frequencies when aligning a series with the Poisson distribution is done in the following order:

is the arithmetic mean, Statistical study of variation; the table determines Statistical study of variation; for each value x the theoretical frequency is determined.

A number of criteria are used in statistics to assess the randomness or materiality of discrepancies between the frequencies of empirical and theoretical distributions.

One of the main criteria for comparing the frequencies of empirical and theoretical distributions is the Pearson consensus criterion (Statistical study of variation – square):

Statistical study of variation,

where Statistical study of variation are the empirical frequencies;

Statistical study of variation – theoretical frequencies.

To assess the proximity of the empirical distribution to the theoretical one, the probability Statistical study of variation of this criterion reaching this value is determined. If Statistical study of variation the > 0.05, then the deviations of the actual frequencies from the theoretical ones are considered random, insignificant. If Statistical study of variation<0.05, then the deviations are significant, and the empirical distribution is fundamentally different from the theoretical one. The values of the tabulation probabilities Statistical study of variation depending on and the Statistical study of variation number of degrees of freedom Statistical study of variation . For the normal distribution Statistical study of variation, for the Poisson curve distribution: Statistical study of variation. Knowing the Statistical study of variationcalculated , we compare it with the tabular (limit). If Statistical study of variation the actual > tabular, then the Statistical study of variation discrepancy between the frequencies of the empirical and theoretical distributions cannot be considered random. If Statistical study of variation the actual < Statistical study of variation tabular, then the discrepancy can be considered random, and the theoretical distribution in question is suitable for describing the empirical distribution.

Romanovsky’s criterion is defined by:

Statistical study of variation,

where Statistical study of variation is the Pearson criterion;

k is the number of units of degrees of freedom.

If this criterion , then the Statistical study of variationdiscrepancies cannot be considered random. If it < 3, then the discrepancy between the empirical and theoretical frequencies can be considered random.

A.N. Kolmogorov proposed a criterion based on a comparison of the distribution of the accumulation of accumulated frequencies (frequencies):

Statistical study of variation,

where d is the maximum difference between the accumulated rates of the empirical and theoretical distribution series, and N is the number of units of the population. If the distribution is given in frequencies, then:

Statistical study of variation,

where D is the maximum difference between the accumulated frequencies of the two distributions.

Актуальное

СЕЛЬСКОЕ ХОЗЯЙСТВО ЯПОНИИ

Сельское хозяйство остается важной отраслью в экономике Японии, хотя...

Современные тенденции развития мирового валютного рынка

Международные валютные рынки являются важнейшими звеньями мировой финансовой системы...

Международное техническое содействие

Разновидностью международной передачи технологий является международное техническое содействие (technical...

МИРОВАЯ ЭКОНОМИКА И ЕЕ ОСОБЕННОСТИ . НА РУБЕЖЕ ВЕКОВ

Облик современной цивилизации коренным образом отличается от того,...

Интеграционные процессы в Содружестве Независимых государств

Содружество Независимых Государств (СНГ) было создано в 1991 г....
Темы

Предложение: понятие, факторы, эластичность

Предложение — это количество товаров или услуг, предложенное для...

Этапы развития мировой валютной системы

До 1914 года международное движение как долгосрочных, так и...

Необходимость регулирования международных экономических отношений на межгосударственном уровне

Многостороннее регулирование мировой торговли на межгосударственном уровне оказывает все...

СЕЛЬСКОЕ ХОЗЯЙСТВО ЯПОНИИ

Сельское хозяйство остается важной отраслью в экономике Японии, хотя...

Местное управление и самоуправление: состояние и концептуальные положения их совершенствования

Управление на местах осуществляется различными органами, которые отличаются друг...

Совокупный личный доход

Чтобы вычислить общую величину личного дохода, необходимо вычесть из...

Определение равновесного выпуска итеративным методом

В стандартной экономической схеме соподчинение цели и средства устанавливается...

Роль платежного баланса в международной торговле

Оценка и сравнительный анализ роли внешней торговли в рамках...
Статьи по теме

Популярные категории

Предыдущая статья
Следующая статья