Contents P4: Apply a range of statistical methods used in business planning for quality, inventory and capacity management .............................................................................................................. 2 P4.1: Statistical methods for business planning:................................................................... 2 P4.2: Measure of variability ................................................................................................. 2 Variability ...................................................................................................................... 2 Application ...................................................................................................................... 2 Statistical process control: ................................................................................................ 3 P4.3: Measures of probability: ............................................................................................. 4 Probability distributions. ................................................................................................. 4 Normal distribution:. ....................................................................................................... 4 Poisson distribution:. ....................................................................................................... 4 P5: Using appropriate charts/tables communicate findings for a number of given variables. ... 6 P5.1. TYPES OF DATA & MEASUREMENT SCALES: NOMINAL, ORDINAL, INTERVAL AND RATIO ........................................................................................................................ 6 P5.2: Different types of charts/tables and diagrams: ............................................................ 7 P5.2.1:
The use of frequency tables, simple tables, pie charts, histograms, frequency
curves and normal curve...................................................................................................... 7 Frequency tables .............................................................................................................. 7 Simple tables. ................................................................................................................... 7 Pie chart: ......................................................................................................................... 8 Historagam:. .................................................................................................................... 9 Frequency curves: .......................................................................................................... 10 Normal curves: .............................................................................................................. 10 P5.2.2: Advantages and disadvantages of different types of methods: ................................ 10 References .............................................................................................................................. 13
1
P4: Apply a range of statistical methods used in business planning for quality, inventory and capacity management P4.1: Statistical methods for business planning: In this article, I will analyze the living standards of Vietnamese households, namely Binh Thuan. From this analysis, I will understand their per capita income, from which to find the methods, the product approach to the appropriate consumer. P4.2: Measure of variability Variability :A measure of variability for a data distribution is a number that propagates the idea of spreading the values in a dataset around the mean. From there, measures of variability helps us understand the distribution of data sets better, faster. They are measures of variability or measures of variation. In comonly case, there are four measures of variability: range, mean, variance and standard deviation Application: I will apply to Binh Thuan’s case Measurement table Variable
Descriptive
Measurement
Age
The length of time that a Year person has lived
Edu 1
Number of school of HH
Year
Total income
Total income of HH in the VND last 12 mouths
Summary of the data set: Mean
SD
Max
Min
Medium
M
Range
od e Age 52.62962
14.03978
96296296
39528923
90
31
2
50
50
62
Inc
91840.26
18578.70
92073.73
91606.78
91840.26
N/
466.94611
om
18757918
19462016
4933455
88181286
18757918
A
5326398
Per
2360.348
1386.457
9672
488
2088
N/
9184
cap
14814815
54071904
e A
ita inc om e According to statistics of Binh Thuan,, we have three main things : Age, Revenue and Per capita income. First, age. The Mean of age is 52.6296296296296 years old, Standard Deviation is 14.0397839528923 years old, the max is 90 years old ( the oldest ), the Min is 31 years old ( the youngest ), Medium and Mode is 50 years old and Range is 62 years old. Next, about income. The Mean is 91840.2618757918 VND, Standard Deviation is 18578.7019462016 VND, The max income is 92073.734933455 VND, The min income is 91606.7888181286 VND, Medium is 91840.2618757918, Range is 466.946115326398 VND and there is no Mode. Finally. Per capita income. The mean is 2360.34814814815 VND, Standard deviation is 1386.45754071904 VND, Max is 9672 VND, Min is 488 VND, Medium is 2088 VND, Range is 9184 VND and there is no Mode too. Statistical process control: Statistical Process Control (SPC) is an industry standard for measuring and controlling quality in the production process. Quality data in the form of measured product or process in real time during production. This data is then plotted on a chart with predefined control limits. Control limits are determined by the capabilities of the process, while the specification limits are determined by the needs of the customer. With real-time SPC you can: Significantly reduce variations and scraps, Improve the science of productivity, Reduce costs, Discover the personality of the hidden 3
process, React immediately to handle change, Make real-time decision on the store floor. All of them to quality management better Data within the control limits indicates that everything is working as expected. Any change in the control limits is likely due to a common cause - natural variations are expected to be part of the process. If the data is out of control, this indicates that the cause can be attributed to the source of the product variation and must change something in the process to correct the problem before the error occurs. . P4.3: Measures of probability: Probability distributions :Probability distribution is a statistical function that describes all possible values and possibilities that a random variable can perform in a given range. This range is between the minimum and maximum possible statistical values, but when the value can be plotted on the probability distribution depends on a number of factors. These factors include the distribution's mean, standard deviation, skewness and kurtosis. For example: Suppose a die is tossed. What is the probability that the die will land on 5? Solution: When a die is tossed, there are 6 possible outcomes represented by: S = { 1, 2, 3, 4, 5, 6 }. Each possible outcome is a random variable (X), and each outcome is equally likely to occur. Thus, we have a uniform distribution. Therefore, the P(X = 5) = 1/6 Normal distribution: A normal distribution, or a bell curve,is a distribution that appears naturally in many situations. For example, the bell curve is seen in tests such as SAT and GRE. Most students will earn a grade point average (C), while smaller students will earn grades B or D. A smaller percentage of students will score F or A. This produces a differential mingle like a bell (hence the nickname). The bell curve is symmetrical. Half of the data will fall to the left of the mean; half will fall right. Poisson distribution: Poisson distribution is a tool that helps predict the probability of certain events occurring when you know the frequency of events. Poisson distribution gives us the probability of a certain number of events occurring over a fixed period of time.
4
For example: The mean value for an event X is 2 times in a day. Find the probability of event X occurring three times in a day. Solution: Mean, m=2 Probability of the event to occur thrice, P(3;2) = e−2 233! = 0.1804465 5 Binomial distribution: A probability function where each value indicates the probability that the probability that a constant probability result occurs in a statistical test will occur a certain number of times in a series of trials. Example: Deadprints from A to E are rolled 50 times. Find the probability of getting "D" exactly 5 times. Solution: Here, n = 50, k = 5, n - k = 45. Probability of success = probability of receiving "D" = s = 15 Therefore, the probability of failure = probability does not get "D" = 1 - s = 45. Inference :The general idea is that statistical inference is to compare specific statistics from observational data sets (ie, mean, standard deviation, difference between subset of data), with differential Provide appropriate reference to evaluate the meaning of those statistics. When different assumptions are met and specific hypotheses about the values of statistics need to be defined, statistical inference can be a powerful approach to deriving conclusions. The science of effective use of existing data or the results obtained for the purpose of testing those hypotheses. Even in a context where a formal test design is not possible, or when the goal is to explore the data, evaluating the meaning can be helpful. Application: The Binh Thuan’ education Table: 5
Observation
Frequency (people)
%
No degree
51
37.77
Primary
47
34.81
Secondary school
13
9.62
High school
11
8.14
College
1
0.74
University
3
2.22
Other
9
6.66
Analysis: Through the above graphs, we find that in Binh Thuan, the heads of households do not have a high level of education. Most of them have no degree or just finished primary school. Some people stop studying in secondaryl and high school only a few have college or university degrees, no PhD or M.S
P5: Using appropriate charts/tables communicate findings for a number of given variables. P5.1. TYPES OF DATA & MEASUREMENT SCALES: NOMINAL, ORDINAL, INTERVAL AND RATIO The first measure is the nominal level of measurement. At this level of measurement, the numbers in the variable are used only to classify the data. At this level of measurement, words, letters and alphanumeric symbols can be used. Suppose there are data on people of three different sex types. In this case, the gender is example, including male and female; Marital status as maried, divorce, widow, code of district as 597, 598, 599 ; or code of region as 8, 7, 12 The second measure is the ordinal level of measurement. This measurement level describes the order relationship between the observations of the variable. In this case, an ordinal is Edu 2, the highest level of education of household head
6
The third measurement is the interval level of measurement. Distance measurement not only classifies and orders measurements, it also indicates that the distance between each time scale on the scale is the same across the scale from low to high. For example, Income and Income per capita is interval The fourth measure is the ratio level of measurement. At this level of measurement, observations, in addition to equal time intervals, can also be zero. Zero in the scale makes this type of measurement unlike other types of measurements. The example is age and number of family members P5.2: Different types of charts/tables and diagrams: P5.2.1: The use of frequency tables, simple tables, pie charts, histograms, frequency curves and normal curve. Frequency tables: In Statistics, it is likely that certain values in the data are collected for repetition. The number of times a specific data value is repeated is called the frequency. The table is based on the data values with its corresponding frequency called the frequency table Example : Table about Binh Thuan’ edu sample Observation
Frequency (people)
%
No degree
51
37.77
Primary
47
34.81
Secondary school
13
9.62
High school
11
8.14
College
1
0.74
University
3
2.22
Other
9
6.66
Simple tables : A table is a data structure that organizes information into rows and columns. It can be used for both archiving and displaying data in structured format. For example, the database stores data in the table so that information can be accessed quickly 7
from specific rows. Example : Table about Binh Thuan’ edu sample Count of Gender Edu2
Total
No degree
51
Primary
47
Secondary school
13
High school
11
College
1
University
3
Other
9
Grand Total
135
Pie chart: Pie charts display data, information, and statistics in easy-to-read 'pie-slice' formatting with different slices of size letting you know how many data elements exist. The larger the slice, the more specific data is collected. Example : Binh Thuan’ edu sample
8
Pie chart for Binh Thuan's education
51
135 47
13 9
11 31
No degree
Primary
Secondary school
High school
College
University
Other
Total
Historagam: A diagram consists of rectangles whose area is proportional to the frequency of a variable and has a width equal to the period of the layer. Historagam of the educational level of the sample
9
Histogram for "Income"
No. of obs.
20 15 10 5 0 91,600 91,650 91,700 91,750 91,800 91,850 91,900 91,950 92,000 92,050 To To To To To To To To To and 91,650 91,700 91,750 91,800 91,850 91,900 91,950 92,000 92,050 over
Frequency curves: a curve with frequency distribution graph Normal curves: A symmetrical bell curve represents the probability density function of the standard distribution. The area of the vertical part of the curve represents the probability that the random variable is located between the section values P5.2.2: Advantages and disadvantages of different types of methods: Frequency tables: Advantages: Information dissemination to the reader is quick and easy, jchis intensive is the trend of information. The frequency tables can help researchers examine the relative abundance of each target data. in their sample. Relative abundance represents the number of data sets that include the target data. Relative abundance is usually expressed as a histogram, but can easily be displayed in the frequency table (Ari Reid,2018) Disadvantages: One disadvantage is that because of the nature of a "table", complex data sets are displayed on a very incomprehensible frequency table. Large datasets can be divided into periods of time for easy display by the frequency table. In addition, the deviation and data may be unclear in the frequency table. (Ari Reid,2018) Simple tables: 10
Advantages: The table allows for "random access", instead of scanning the whole list or segment. So if the information is such that a reader will only want a part of it, a table has the advantage. For example, consider tuition at a university, depending on the student's credit score relative to the graduating student, versus the graduating student, in the state versus out-of-state. It is likely that the reader just wants to find out the fees associated with their own condition. (JRAINEHARRISON, 2018) The second case in which a table has the advantage is that when the reader wants to compare information between rows or columns. For example, you have benchmark results for some systems. If rows are different criteria and columns are different systems, the reader can easily compare how different systems have performed on the same benchmark (row scan) or how one Single system implemented on different standards (scan column). It would be tedious to make correlations with the data in a list or paragraph. Disadvantages: lots of coding and the need to keep track of the number of rows and columns so some data can span over multiple.Next, not so responsive, divs are easier (JRAINEHARRISON, 2018 ) Pie chart: Advantages: Pie charts show simple and easy-to-understand data. It can be an effective communication tool for even an unknown audience, because it represents visual data as a fraction of a whole. Readers or audiences see a data comparison at a glance, allowing them to make an immediate analysis or to understand the information quickly. This kind of data visualization chart eliminates the need for the reader to self-check or measure basic numbers. You can also manipulate pieces of data in a circle Disadvantages: The pie chart becomes less efficient if it uses too much data. For example, a chart with four slices is easy to read; one with more than 10 becomes less, especially if it contains many slices of the same size. Adding labels and data figures may not be useful here, as they can become crowded and hard to read by themselves. This chart type represents only one dataset - you need a string to compare multiple sets. This can make it 11
harder for the reader to analyze and assimilate information more quickly. Comparing the data slices in a circle also has its problems, because the reader must take into account the angles and compare the non-contiguous slices. Data manipulation in the design of a chart can cause the reader to draw inaccurate conclusions or make decisions based on visual impact rather than data analysis Historagam: Advantages: Although charts are considered some of the most commonly used charts to display data, the chart has many advantages and disadvantages hidden in its formula settings. Charts allow viewers to easily compare data and, in addition, they work well with a large range of information. They also provide a more specific consistency, since time intervals are always equal, an element that allows easy transfer of data from frequency tables to charts. Although useful in many different cases, charts are especially useful when handling large value ranges. For example, if a sample of college students consists of more than 2000 students and we are collecting data about the number of times someone has driven their car outside the campus over the past year, the range may differ from People have yet to find a reason to drive their car outside the campus for an individual who feels the need to visit home every week. The range of data values will be very large, so in this case, it is convenient to use the chart. Disadvantages: Although there are many cases where using the chart is considered convenient, there are also cases where using or interpreting a chart can be troublesome. For example, when interpreting a chart, it is extremely difficult and practically impossible to extract the exact "input" quantity in the chart unless it is a histogram. For example, if you were given a chart and asked how many people had given their data in a survey, it would be extremely difficult to determine an exact number. Charts are often considered inconvenient when comparing multiple categories, because although you can compare several charts side-by-side, it does not produce the desired effect. Frequency curve: Advantages: The frequency curve has the greatest advantage when displaying the 12
deviation of the distribution whether it is deflected, negatively distributed, and symmetric. Disadvantages: Frequency curves have a major disadvantage when not displaying the exact values of the distribution. It is also difficult to compare different data sets. Normal curve: Advantages : It is always important to identify top performers and reward them with respect. This helps the organization retain top talent with them. In this way, both employees and organizations develop. Normal curves are probably the only method by which managers in the organization can perform tolerance or rigorous ranking. If managers exercise leniency in their rankings, most employees fall into the high ranking category. If managers perform a rigorous level of their rankings, most employees fall into the lower rank category, which will result in loss of human resources. So, on average most managers tend to evaluate on a tolerant scale. Disadvantages : The Bell curve model can become too rigid in case the strength of the staff in the organization is less. Here managers may be forced to put employees into specific ratings just for the benefit of bell curve requirements. As in the bell curve model, the manager can only provide a limited number of employees in the top performing category, the employee who actually performs well in the year may be forced to be classified in the name Average performer item for a number of valid Bell curve requests. This will lead to dismay among employees.
References Anon., 2018. Collinsdictionary. [Online] Available at: https://www.collinsdictionary.com/dictionary/english/normal-curve Anon., 2018. Enotes. [Online] Available at: https://www.enotes.com/homework-help/can-someone-help-me-with-this13
late-already-have-440445 Anon., 2018. Enotes. [Online] Available at: https://www.enotes.com/homework-help/can-someone-help-me-with-thislate-already-have-440445 Anon., 2018. Infinityqs. [Online] Available at: https://www.infinityqs.com/resources/what-is-spc Anon., 2018. Investopedia. [Online] Available at: https://www.investopedia.com/terms/p/probabilitydistribution.asp Anon., 2018. NCBI. [Online] Available at: https://www.ncbi.nlm.nih.gov/books/NBK299417/table/findings.t4/ Anon., 2018. techterms. [Online] Available at: https://techterms.com/definition/table [Accessed 6 6 2011]. Bill, 2018. Answers. [Online] Available
at:
http://qa.answers.com/Q/What_are_the_advantages_and_disadvantages_of_frequency_c urve CavinHR, 2018. CavinHR. [Online] Available at: http://www.cavinhr.com/relevance-of-bell-curve-method-of-performanceappraisal/ [Accessed 8 10 2014]. Finch, C., 2018. bizfluent. [Online] Available
at:
https://bizfluent.com/list-6715678-advantages-disadvantages-pie-
chart.html [Accessed 22 5 2018]. Finch, C., 2018. bizfluent. [Online] Available
at:
https://bizfluent.com/list-6715678-advantages-disadvantages-pie-
chart.html [Accessed 22 5 2018]. Kenney, J. F. a. K. E. S., 2018. Mathworld. [Online] Available at: http://mathworld.wolfram.com/FrequencyCurve.html 14
O'Shea, M., 2018. Quora. [Online] Available at: https://www.quora.com/What-are-the-advantages-and-disadvantages-ofusing-Database-views [Accessed 11 4 2018]. Reid, A., 2018. Sciencing. [Online] Available
at:
https://sciencing.com/advantages-disadvantages-frequency-table-
12000027.html [Accessed 13 3 2018].
15