Question :
Business data analyst of an International Business consulting company has to make a report which will concentrate on utilisation of statistical methods for decision making process.
- Analyse business and economic information that is available on published sources.
- Critically evaluate raw business data by applying different statistical methods.
- Apply various statistical methods at the time of business planning.
- Use suitable charts and tables for communicating findings.
Answer :
INTRODUCTION
Business and economic data analysis is the process of collecting useful information in quantitative and qualitative form. It is the collection of data, statistics and figures to interpret and analyse it into useful information for the analysts to take economic decisions and for the business planning (Hong-ling, 2012). The analysis of business and economic data allows the person to answer certain question and take decisions regarding investment activities.
TASK P1 Nature and process of business and economic data:
This can be said that the descriptive statistics would be used by the organisation for summarising and evaluating data, confidence intervals and hypothesis testing. The business managers is required to have statistical model relied upon the statistical model based decision support systems (Chelladurai and Kerwin, 2017). Statistical skills effectively help the organisation to gather, assess and evaluate the data which is interpreting the data relevant to their decision making. statistical thinking accessing to resolve the problems in a diversity of contexts and also to add substance to the decisions.
P2 Evaluate data from variety of sources:
MEASURES OF CENTRAL TENDENCIES:
YEAR |
SELLING EXPENSES |
NET INCOME |
2009 |
15000 |
20000 |
2010 |
20000 |
25000 |
2011 |
24000 |
30000 |
2012 |
20000 |
40000 |
2013 |
18000 |
35000 |
2014 |
25000 |
50000 |
2015 |
20000 |
45000 |
2016 |
30000 |
65000 |
2017 |
24000 |
45000 |
MEAN |
21777.78 |
39444.44 |
MEDIAN |
20000.00 |
40000.00 |
MODE |
196000 |
355000 |
Descriptive Statistics
|
|||||||||
|
N |
Range |
Minimum |
Maximum |
Mean |
Std. Deviation |
Variance |
||
Statistic |
Statistic |
Statistic |
Statistic |
Statistic |
Std. Error |
Statistic |
Statistic |
||
VAR00003 |
9 |
15000 |
15000 |
30000 |
21777.78 |
1479.281 |
4437.842 |
19694444.444 |
|
VAR00004 |
9 |
45000 |
20000 |
65000 |
39444.44 |
4598.040 |
13794.121 |
190277777.778 |
|
Valid N (listwise) |
9 |
|
|
|
|
|
|
|
|
Statistics
|
|
|||
|
VAR00003 |
VAR00004 |
||
N |
Valid |
9 |
9 |
|
Missing |
1 |
1 |
||
Mean |
21777.78 |
39444.44 |
||
Median |
20000.00 |
40000.00 |
||
Mode |
20000 |
45000 |
||
Sum |
196000 |
355000 |
||
VAR00003 |
|
|||||
|
Frequency |
Percent |
Valid Percent |
Cumulative Percent |
||
Valid |
15000 |
1 |
10.0 |
11.1 |
11.1 |
|
18000 |
1 |
10.0 |
11.1 |
22.2 |
||
20000 |
3 |
30.0 |
33.3 |
55.6 |
||
24000 |
2 |
20.0 |
22.2 |
77.8 |
||
25000 |
1 |
10.0 |
11.1 |
88.9 |
||
30000 |
1 |
10.0 |
11.1 |
100.0 |
||
Total |
9 |
90.0 |
100.0 |
|
||
Missing |
System |
1 |
10.0 |
|
|
|
Total |
10 |
100.0 |
|
|
||
VAR00004 |
|
|||||
|
Frequency |
Percent |
Valid Percent |
Cumulative Percent |
||
Valid |
20000 |
1 |
10.0 |
11.1 |
11.1 |
|
25000 |
1 |
10.0 |
11.1 |
22.2 |
||
30000 |
1 |
10.0 |
11.1 |
33.3 |
||
35000 |
1 |
10.0 |
11.1 |
44.4 |
||
40000 |
1 |
10.0 |
11.1 |
55.6 |
||
45000 |
2 |
20.0 |
22.2 |
77.8 |
||
50000 |
1 |
10.0 |
11.1 |
88.9 |
||
65000 |
1 |
10.0 |
11.1 |
100.0 |
||
Total |
9 |
90.0 |
100.0 |
|
||
Missing |
System |
1 |
10.0 |
|
|
|
Total |
10 |
100.0 |
|
|
||
MEAN: Mean or average is the most common measure of central tendency for analysis of statements. It is calculated as the sum total of all the values in a data to be used and dividing it by the total number of the values in the data (i.e. n) . In the above example the mean is calculated by taking the sum of all the selling expenses and net income which is then divided by the number of years that data is presented for. For example, mean of selling expenses during the period was 196000/9= 21777.78 and mean of net earnings was calculated as 355000/9=39444.44 which is shown in the table above.
MEDIAN: Median is calculated by re-arranging the data in the ascending or descending order and then choosing the middle number of the data, that number is called the median of the data (Northam, 2012). Median is calculated by taking the number of years in a data (which is denoted as 'n') and dividing that number by 2, that value will be median of data(n/2). The median in the above data in case of selling expenses is 20000 and in case of net income the median is 40000.
MODE: Mode is that measure of central tendency which occurs in the highest frequency in the given data set. In the above example the modal value of selling expenses is 20000 and mode of net income is 45000.
MEASURES OF VARIABILITY:
Standard deviation: This measure of variability calculated the variation and dispersion of the data set. A low standard deviation indicates that the data points are close to mean of the given data set and a standard deviation which is higher in number shows that the mean of the data is dispersed or spread over a wide range. In the above data set the standard deviation of selling expenses in 4437.842 and the standard deviation of net income is 13794.121.
Variance: The variance indicates the difference between the actual result and the expected result, for example a difference between a budget made by a company and the actual expenditure it incurred. It is used in the statistics of probability distribution; it measures the spread between the numbers in the data set. Variance is used in Monte Carlo simulation, hypotheses testing etc. The variance of selling expenses and net income in the above data set is 19694444.444 and 190277777.778 respectively.
Exploratory analysis: Exploratory data analysis is an approach to summarise the important characteristics of a data. EDA employs a lot of techniques for the analysis of data to maximise the insight into a data set, it uncovers the underlying structure, extraction of important variables, detection of outliers and anomalies, testing the assumptions that underlies etc (Venables and Ripley, 2013). It is precisely an approach and not a form of technique. EDA is a philosophy for the dissection of a data, what we look into a data, how we look at it, how we should interpret the data. EDA uses certain types of techniques for plotting the raw data such as data traces, histogram, Bi-histograms, lag plots. Plotting statistics such as mean plots, standard deviation etc.
Confirmatory data analysis: Confirmatory analysis is most commonly used in in social research. The main purpose of confirmatory factor analysis is is to find whether the data fits into the hypothesised measurement model. This hypothesis model is based on the previous data for analytic research that is conducted. It is the data analysis that is used for the confirmation or rejection of a measurement theory. The main benefits of confirmatory analysis are that the results are more meaningful, the main reason behind this research is that it reduces the probability of false reporting of a coincidental research as a meaningful. This is called the probability of a type 1 error.
P3 Methods of data analysis:
Qualitative data analysis: This analysis of data is generally non-numerical and non-statistical. It is generally the interpretation and understanding of underlying reasons, opinions and motivations (Neave, 2013). This kind of data is available in manager discussions and analysis. This is the non-statistical methodological approach that are mainly observed by the concrete material at hand. This also measure the hope of developing various kinds of rules and norms at the time of qualitative research which can be explained as the investigation of what is presumed to be the dynamic reality. Basically, this can be rightly said that the qualitative research is the huge, investigated and valid procedures that could contribute to detailed knowing of the context. This is the main tool by which the data which is in the qualitative form is analysed in an effective manner which ultimately helps to make certain tools better and effective. Here, the interpretation is based on the basis of graphs and charts which could ultimately help out to make the reliable conclusion for the business.
Quantitative data analysis: This analysis is a numerical study of data to interpret various important information about the company. This analysis uses statistical methods such as measures of central tendencies and measures of dispersion to derive results (Linoff and Berry, 2011). It uses financial statements such as profit and loss account, balance sheet, and cash flow statements to derive the numeric to be used in analysis. This is simply the use of quantitative and statistical tools in order to evaluate the investment opportunities and form the appropriate decisions. Under this research method, quantitative or numerical work likewise helps to explain the distribution of the third sector firms, evaluate their contribution to the community and economy and evaluate their dynamic manner.
Criteria |
Qualitative |
Quantitative |
Purpose |
Identify and evaluate social interactions |
Test hypothesis, test cause and effect. Develop projection for the forthcoming time period. |
Studied group |
Measure selected intentionally. |
Additional and selected randomly. |
Data type |
Here the data is gathered by way of words, image, aims. |
Numbers and statistics |
Data form |
Here are various kinds of tools through which data can be gathered via: Open ended reactions, participant observations, field notes interviews. |
In this various quantitative data is gathered via: Conscious calculations by adopting structures and reliability device for data collection. |
Type of data assessment |
Patterns, features, themes determination. |
Statistical connection identification. |
Researcher’s role |
Researcher might be identify in the assessment and participant’s characteristics might be recognised to the researchers. |
Researcher and their biases which are not recognised to participants in the assessment. Participant characteristics are not disclosed. |
Results |
Specific findings, lower generalizable. |
Common findings could be implemented to the other populations. |
There are so basically two kinds of sites which can be ideas of organisation statistics which is frequent and appropriate manner. There are different methods of analysis for the interpretation of the data from the various sources which are discussed below:
Descriptive Analysis: Descriptive analysis of data is concerned with the usage of past data by processing the useful information that is to be used and summarising and processing that data to be used in the further analysis. This analysis uses descriptive statistics such as measures of central tendencies (mean, median, mode) and measures of variability (standard deviation and variance).
This is numerical statistical data which must be reflected preciously and also in such a way so that the decision maker could be quickly gain from the important characteristics of the data for making them into the decision process (Horton, Baumer and Wickham, 2015).
Descriptive quantity is emerged from the sample data is the average which is also known as the arthimatic average of the sample data. This serves as the highly valid single calculate of the value of a typical member of the sample. If sample covers some values which is big or small which have an exaggerated effect on value of the mean, sample is highly accurate which is characterized by the median which simply reflects the middle numbers of the set of observation.
Inferential statistics: This is concerned with form inferences from samples about the populations from which they are drawn. On the other hand, if there is a difference between two samples, then there is a need to know the real difference of just a chance difference. Inferential statistics can be implemented for elaborating a phenomenon for reliability of a claim. Under this circumstances, inferential statistics is known as the exploratory data analysis.
Which are discussed below:
P4 Statistical methods used in business planning for quality, inventory and capacity management:
Statistical process control tools extend implemented of descriptive statistics in order to control quality product and process. By implementing statistical process control in order to identify amount of variation which is common. After that, there is a need to control the production process in order to assure that the manufacturing stays within the normal range. The most famous tool for controlling the manufacturing process which is a control chart. There are diverse kinds of control chart which are implemented to control diverse kind of manufacturing process.
Emerging control Charts:
A control chart is known as the graph which reflects whether a sample of the data limits within the normal range of variation. Control chart have upper and lower control limits which segregate common from assignable causes of variation. The common range of variation is elaborated by implementing control chart limits. The table shows control chart for the Cocoa Fizz bottling operation. X axis reflects samples which have taken from the process over the time. Y axis reflects the quality characteristics which is being monitored.
Control charts for variables: This charts for variables control characteristics which could be evaluated and have regular scale like height, weight, volume or width. If item is inspected, then the variable is being controlled and recorded (Dey, MüIler and Sinha, 2012). Two of the key common important used control charts for variables control both central tendencies of the data and variability of the data.
Mean (x-bar) Charts: Mean control chart is normally known as an x-bar chart. This has normally been control changes during mean process. To incorporate a mean chart, there is a requirement to form the centre line of the chart. To make this, multiple samples and measure their means. Normally these samples are small, with about four or five observations. Each sample has its own mean. Centre line of the chart is then measured as the mean of all x sample means, where x is the number of samples.