JPS62297959A - Electronic calculator - Google Patents

Abstract

PURPOSE:To perform a distribution functional operation in a short time with no deterioration of accuracy by providing a distribution functional operation instruction key to an electronic calculator together with a numerical key which supplies the variable value such as the probability value, the freedom degree value, etc., and a nonvolatile memory device which stores a distribution functional operation routine. CONSTITUTION:The character keys define by the group of distribution functional keys and used for statistic processing with the operation of a distribution function designation key 12A are arranged to keys set which is at the lower stage part of a group 12 of character input keys. The arithmetic formula of distribution function is generally known in the statistics and the arithmetic processing routine formed by said arithmetic formula is stored in a nonvolatile way. An approximation formula is used with a large freedom degree and a recurrence formula is used with a small freedom degree. At the same time, only the result on the recurrence formula of the part of the upper probavility (p) having a high using a frequency is stored to increase the calculation speed. Thus it is possible to perform the distribution functional operation without referencing a book like a numerical table, etc., for the execution of a statistic examination requiring a distribution function or estimation of sections.

Description

[Detailed Description of the Invention] 3. Detailed Description of the Invention (Field of Industrial Application) The present invention relates to a small-sized electronic computer with a distribution function function.

<Prior art> Conventionally, when calculating a distribution function on a small electronic computer, a RASIC program was created based on approximate formulas and recurrence formulas known from the "Statistical Numerical Tables" published by the Japanese Standards Association. I was doing calculations.

In creating this program, an approximation formula is used to calculate the distribution, but because the approximation formula has parts with poor calculation accuracy, a recurrence formula is used to calculate these parts.

In addition, in the case of program calculations that use a distribution table that summarizes the calculation results of recurrence formulas, the distribution table itself is stored in a storage device,
Calculations are performed by sequentially calling and using applicable values.

<Problems to be solved by the invention> However, calculations using recurrence formulas generally require a lot of time, and storing distribution tables in a storage device requires a large storage capacity. It is impossible to cover the entire range of

The present invention was made in view of the above-mentioned problems.
By compensating for the mutual shortcomings of the above-mentioned operations, distribution function operations can be performed over a wide range in a short time without loss of accuracy.
The purpose is to provide a small electronic computer that is equipped with a function that is linked to the computer.

Means for Solving the Problems> The compact electronic computer of the present invention has a distribution function calculation instruction key, a numeric key for inputting variable values such as probability values and degree of freedom values, and a distribution function calculation instruction key based on the distribution function calculation instruction key. and a non-volatile storage device that stores functional operation routines.

Function> In the small electronic computer of the present invention, distribution function calculation is performed according to the distribution function calculation routine stored in the storage device based on the input of the distribution function calculation instruction key and variable values such as probability values and degree of freedom values. It is something to do.

<Example> A detailed explanation will be given below based on the drawings.

FIG. 1 is a diagram showing the appearance of a small-sized computer according to an embodiment of the present invention, and this small-sized computer is designed to be compact so that it can be held and operated with one hand.

Reference numeral 1 denotes a small computer main body, and a liquid crystal display device (hereinafter referred to as LCD) 2 is provided in the upper left corner, and a keyboard having character input keys and function keys is provided around it. The LCD 2 is of a dot segment type and has a display capacity for one line of, for example, 24 digits, and further includes indicators 21, 22, .
23.24 are provided.

Reference numeral 10 denotes a power switch, and by turning on this power switch, the main body 1 is powered on and initialization processing is performed. This initialization presets the mode of the main body to statistical processing mode (STAT).

11 is a mode setting key, which is a statistical processing mode (5TA
The 5TAT key 11A is used to alternately set the calculation mode (CAL) and the calculation mode (CAL) for each operation.
It consists of a RAS IC key 11B which is set alternately for each operation of uN) mode and programming mode (PRO).

Indicator 5TAT21 is lit in the above-mentioned 5TAT mode, indicator CAL22 is lit in CAL mode, and indicator RuN2 is lit in RuN mode.
3 lights up and the indicator PR024 is in PRO mode.
lights up.

Reference numeral 12 denotes a group of character keys, which can also be defined as various function keys. Particularly in the lower part of the keys, character keys are arranged which are defined as a group of distribution function keys used for statistical processing by operating the distribution function designation key 12A.

Here, a group of these distribution function keys will be explained.

N(p) assigned to character Z key 1211 character
) key and N(x) key are the operation instruction keys for the normal distribution function, and the N(p) key is the random variable X for the upper probability P.
The N (T) key is a calculation instruction key that instructs the calculation of a probability p higher than the probability variable X.

The t(p) and t(x) keys respectively assigned to the letter C key 123 and the letter Y key 124 are used when dealing with problems such as testing and estimation of the average value, or testing and estimation of the difference between two average values. This is a calculation instruction key for the t distribution function, and t (p
) instructs calculation of a random variable

The x2 (p) and X2 (x) keys respectively assigned to the letter B key 1252 and the letter N key 126 are used to test for variations.

This is the calculation instruction key for the X2 distribution function when dealing with estimation or goodness-of-fit testing. X2(p) instructs the calculation of the random variable
x) is an instruction key for instructing the calculation of the higher probability p than the degree of freedom n and the random variable X.

Letter M key 127. The F(p) and F(transmission) keys respectively assigned to the space (spc) key 128 are keys for calculating the F distribution function when dealing with the same degree of variation, and Fφ) is for degrees of freedom n1. Instructs calculation of random variable X for n2 and upper probability p, F(x) has degrees of freedom nl+
This is an instruction key for instructing the calculation of the higher probability p from n2 and the random variable X.

The B key assigned to the letter J key 129 is a calculation instruction key for the binomial distribution, that is, the B distribution function, and instructs the calculation of the probability that an event with probability p will occur X times out of n.
The p key assigned to is a calculation instruction key for a Poisson distribution function, that is, a 2-distribution, and is an instruction key for instructing calculation of the probability that an event with an average of m will occur X times.

The arithmetic expressions for these distribution functions described above are generally known in statistics and the like, and the small electronic computer of the present invention non-volatilely stores arithmetic processing routines based on these arithmetic expressions.

Numeral 13 is a number entry key and four arithmetic operation keys, and 14 is a group of operation instruction keys for obtaining average values and deviation values in statistical processing.

15 is a cursor mark movement key of the display device;
is an execution key that instructs to execute a calculation based on a calculation instruction.

FIG. 2 is a circuit block diagram of a small electronic computer according to the present invention.

40 is a CPU, which is connected to R via a path line.
Data processing is executed according to various control programs stored in the OM45.

This ROM 45 includes an area 451 for storing an initial program that controls the initial setting process when the power is turned on, and
An interpreter area 452 for the ACIC language, an area 453 for storing an arithmetic processing routine based on an arithmetic expression of a distribution function, and a dot for outputting (in this embodiment, for displaying) code information of each character and symbol generated by the CPU 40. Character pattern generator (CG) that develops into patterns
An area 454, an area 456 for storing various other system programs, and a normal distribution are distributed.

A distribution table 457 area (each data value of this distribution table is generally known as a distribution table) that stores calculation results based on child-determined conditions of distribution functions such as x2 distribution and 2 distribution as a table; It is equipped with

It is preferable that these areas are each formed by separate ROM chips.

41 is the above-mentioned LCD which is driven for display by a display control section 42 connected to the CPU 40. This display control section 42 has a built-in display buffer that stores display data (dot data).

44 is the keyboard device shown in FIG. 1, which supplies coded signals corresponding to operated keys to the CPU.

46 is a RAM read directly by the CPU via a pass line, similar to the ROM 45, and is a RAM arbitrarily created by the operator.
Area 461 for storing user programs based on the SIC language and various buffers.

It also includes a system area 462 allocated to counters and the like.

Figure 3 shows the methods for determining the distribution function, namely (a) method for determining using an approximation formula (t-) method for determining using a recurrence formula (→ (i) method for storing the distribution table in a storage device) Accuracy of the obtained value (ii) Required storage capacity Taking as an example, cases where the degree of freedom n is large (n≧40) and cases where the degree of freedom is small (n(40)) indicate suitability (indicated by 0) and unsuitability (indicated by X).

As is clear from the figure, when the degree of freedom is large, highly accurate values can be obtained even by using approximate expressions.

However, when the degrees of freedom are small, approximate expressions have poor accuracy, so it is necessary to use recurrence expressions that require a long calculation time. Furthermore, it is impossible to store all the calculation results of the recurrence formula in the range in which the recurrence formula must be used due to capacity limitations.

However, in a function (hereinafter referred to as t(n, p)) that calculates the value t of a distribution point from the upper probability p of the t distribution, the parameters are the degree of freedom n and the quality probability p.
There are several items that are used very frequently.

Therefore, in the present invention, when the degree of freedom is large, an approximation formula is used,
When the degree of freedom is small, the calculation speed is increased by using a recurrence formula and storing only the calculation results of the recurrence formula for the frequently used portion of the upper probability p.

The frequently occurring upper probabilities p are set to 5% and 1%, and the calculation results of the recurrence formula for these parts are stored in the ROM 45.

FIG. 5 is a flowchart showing the procedure for determining which method to use for calculation at t(n,p).

First, it is determined whether the degrees of freedom are 40 or more (Sl)
, if the degree of freedom is 40 or more, the calculation is performed using an approximate formula (
S2), display the calculation result on the display device (S3)0 Also, if the degrees of freedom are 39 or less, the upper probability p is R
It is determined whether it is stored in the OM 45 (
S4), and if the value is stored, a calculation is performed using that value (S5), and the calculation result is displayed on a display device (S5).
3).

If the upper probability p is not stored in the ROM 45, calculation is performed using the recurrence formula (S6), and the calculation result is displayed on the display device (S3).

In addition, each calculation sequence in each distribution function key is RO
It is semi-permanently stored in the statistical processing program in M45. Each calculation sequence is processed according to the flowchart shown in FIG. 5 described above. In other words, each calculation of the normal distribution function, the x2 distribution function, and the F distribution function can be performed based on an approximation formula, a recurrence formula, or It is determined whether the calculation is to be performed or not based on a predetermined statistical value, and the calculation is performed. The calculation result is supplied to the display device and displayed. Also, corresponding to the distribution function calculation keys, commands in the RASIC language as shown in FIG. 7 are similarly provided, and based on this command, Statistical calculations are possible by creating a user program. This arithmetic sequence utilizes a key-based arithmetic routine, and the results can be displayed in the same manner as described above or stored in specific separate registers.

Next, a method of operating a distribution function calculation in a small electronic computer will be explained. There are two methods for this operation. That is, I) A method of using it as a function that can be used by linking with a RASIC program.

口）When you press the key defined for a distribution function, a variable input message is displayed on the display, and when you input numerical values according to the message, the value of the function is obtained and displayed on the display.

As an example of O (2), each operation method will be explained in detail below, in a t distribution with n degrees of freedom, the upper probability p
A case will be explained in which the function t(n, p) determined by is defined as PTD(n, p). RAS I small electronic computer
When inputting "PTD (9,001) E 下工亚" as the C input state, the value 2.821 when the upper probability is 1% in the t distribution with 9 degrees of freedom is displayed on the display.

Also, during the RASIC program, X=PTD(9,0,01) Then, the value 2.821 mentioned above is assigned to X.

This can be written as follows to obtain the same result: 0 N=9 :P=1 :X=PTD(N, P/100)
In the case of the example of ■), the t distribution function PTD (
n, p) are assigned to the letter C key 123, and calculations are performed by operating the assigned t (p) key.

That is, the small electronic computer is set to 5TAT mode for executing a calculator-like statistical calculation function. When the t (p) key is input after the distribution function key 12A in this 5TAT mode, a sequence is entered to obtain the distribution function PTD(n,p). Specifically, the operation for determining pTD(9, o, ol) is shown in steps S11 to StS in FIG.

Sit is in a state of waiting for the input of the degree of freedom n in the t distribution due to the input of the distribution function key t (p) key, and S1□ indicates the stage where 9 is input as the value of n. . S13 is in a state where the input of the upper probability P is waited for as the next variable because the key is input after 9, and S14 is 0.01 as the value of P.
This shows the state in which . Then, in 815, 0.
Since the <OK> cause key is input after 01, 2.821 is obtained as the value of the distribution function value PTD (9, 0, 01) and is displayed. As is clear from the above, in this embodiment, when calculating the distribution function, a display prompting the input of necessary variables is made, so the operation is simple.

FIG. 8 is a flowchart illustrating the mode state setting operation of the small computer of the present invention.

First, when power is input (S21), initialization is performed by executing the initialization program stored in the ROM 45 (S22), and the 5TAT symbol 2
1 to turn on and set it to statistical processing mode (S23) O(
The display is zonal. ) In this state, the mode key 11 is operated to read a key to select the RASIC processing mode or the calculation mode (CAI, mode) (S24).

When the distribution function instruction key 12A is operated here, the arithmetic processing specified by the distribution function key group 12B described above is executed (S26).
27), the BASIC processing mode is entered, and RASIC program processing is performed (328) o. Then, in BASIC mode, by operating the RASIC key again, the program creation mode (PRO) and program execution mode (RUN) can be selected. It will be done.

Furthermore, by operating the 5TAT key (829
)CAL mode, and the corresponding mode is set by operating another key.

<Effects of the Invention> The small electronic computer of the present invention has a distribution function calculation instruction key,
It is equipped with numeric keys for inputting variable values such as probability values and degrees of freedom values, and a non-volatile storage device for storing distribution function calculation routines based on the distribution function calculation instruction keys, so it can be used for statistics that require distribution functions. It is an extremely convenient small electronic computer because it can perform calculations without referring to books such as numerical tables when performing statistical tests and interval estimation.

In addition, the distribution function calculation routine can be performed by selecting one of the following: an approximate formula calculation routine, a recurrence formula calculation routine, or a calculation routine that selects and uses the corresponding calculated value from a fixedly stored group of calculated values. In order to do this, the required calculation results are highly accurate and can be obtained in a short time, and the storage capacity is small, so a small electronic computer with extremely high calculation capacity is used.

[Brief explanation of the drawing]

Figure 1 is an external view of a small electronic computer according to the present invention, Figure 2 is a circuit block diagram of the computer, Figure 3 is a diagram showing the suitability and unsuitability of methods for determining distribution functions, and Figure 4 is a diagram of the computer. FIG. 5 is a flowchart showing the operation of selecting the calculation method of the computer; FIG. 6 is a diagram showing the specific calculation operations of the computer; FIG. 7 8 is a diagram showing RAS IC language commands corresponding to distribution function calculation keys, and FIG. 8 is a flowchart showing mode setting operation of the computer. 11; Mode setting key, 12: Group of letters, 12A
: Distribution function designation key, 121 to 130: Character input keys which also serve as distribution function calculation instruction keys, 113: Number input key and four arithmetic operation keys, 14: Group of calculation instruction keys, 16: Execution key, 40: CPU. 45: ROM, 46: RAM. Agent Patent Attorney Takeshi Sugiyama (and 1 other person) Figure 2 Figure 3 Genus 4 Figure

Claims (1)

[Claims] 1. A distribution function calculation instruction key, a numeric key for inputting variable values such as probability values and degrees of freedom values, and a non-volatile memory that stores a distribution function calculation routine based on the distribution function calculation instruction key. A small electronic computer characterized by comprising a storage device and performing distribution function calculations. 2. The distribution function calculation routine described above can be performed by selecting one of the following: an approximation formula calculation routine, a recurrence formula calculation routine, and a calculation routine that selects and uses a corresponding calculated value from a group of calculated values fixedly stored in advance. A small electronic computer according to claim 1, characterized in that the computer performs the following.