стр. 65 |

вЂњRandom Number Generation.вЂќ

Number of Variables: Number of columns of values you want in the output

table. If you do not enter a number, all columns in the output will be

filled.

Figure 158: Random Number Generator

If you do not see this option, then use TOOLS / ADD-INS to activate the Add-In for

34

data analysis. Refer to section 41.4.

230

Chapter 13: Other Tools for Statistics

Number of Random Numbers: Number of data points you want to see.

Each point appears in a row of the output table. If you do not enter a

number, all rows in the output range will be filled.

Distribution: choose the Density Function for defining the criterion for the

Random Number generation.

Parameters: The base parameters for the generation process using the

selected Density Function.

Figure 159: Choice of Density Functions

Table 39: Choice of Density Functions

Distribution Comment on setting parameters for random number generation

Figure 160: Bernoulli

This Density Function is

Bernoulli characterized by a probability of

success (p value) on any given

trial/observation.

Figure 161: Binomial

This Density Function is

characterized by a probability of

Binomial

success (p value) in any one

trial for a number of trials.

Discrete

Figure 163: Discrete Or Custom

231

Statistical Analysis with Excel

Distribution Comment on setting parameters for random number generation

Density Function

Or Custom The range must contain two

Density columns: The left column contains

Function values, and the right column

contains probabilities associated

with the value in that row. The

sum of the probabilities must be 1.

Note: You can use the function

FREQUENCY (A1, A:A)/count

(A:A) to generate the probability

you see in column B.

Figure 162: Parameters

This Density Function is characterized by a mean and a standard

Normal

deviation.

Figure 164: Patterned

Patterned

Figure 165: Poisson

This Density Function is

Poisson

characterized by a value

lambda, equal to (1/mean).

232

Chapter 13: Other Tools for Statistics

Distribution Comment on setting parameters for random number generation

This distributing is characterized by lower and upper bounds.

Uniform Excel draws variables from all values in the range. The

probability of drawing a value is equal for all values in the range.

TIME SERIES

13.3

Exponential Smoothing

This tool and its formula predict a value based on the forecast for the

prior period, adjusted for the error in that prior forecast. The tool uses

the smoothing constant alpha, the magnitude of which determines how

стр. 65 |