Business Statistics Methods - how to Organize and Manipulate the Data

Résumé du document

In the business world, statistics is the main tool to identify, understand and summarize the current situation. In fact, it is the way for a company to change its strategy and to be able to know what it should improve to increase its sales or to answer better to the customers' expectations. In order to be able to do the statistics, we have to collect enough reliable quantitative data. The initial data is, at the beginning, unintelligible, disorganized and unusable. The statistics will summarize this data that will allow us to analyze, compare, interpret and visually present them. It is the main support to draw conclusions. Before the arrival of new technologies, it was very complicated to use this data, but due to the arrival of computers in the 20th century, it is much more easier to sort the information, it takes less time and the results are reliable. In this case study, I am going to explain how to organize and manipulate data in order to get the information we need to know.

Extraits

[...] Usually, we have to select a random values and numbers. If you use a simple random sampling, each of the values will be assumed to come from a uniform probability distribution, which has a maximum value of 1 and a minimum of 0. (Exponential (Times between arrivals / services) It is a class of continuous probability distributions. It goal is to determine the time interval between successive random events: your arrival and the execution of what you came for. For instance when you wait for the bus, the exponential distribution determines the interval before the next bus arrival. [...]

[...] In the first series, the variance is very high: there is an important spread between the two extremes. The other two series have a little spread, which explains why the variance is low . b. Compare the three series by pairs, I mean vs vs vs 3. Can we reject with a significance level of the null hypothesis H0 that the means of both series are the same? (Comparison: 1 and 2 - The mean is quite the same 10,087 10,105. - The variance of series 1 is much higher than the one of series 56,7>1,6. [...]

[...] Before the arrival of new technologies, it was very complicated to use these data: thanks to the computer arrival in the 20th century, it is much easiest to sort the information, it takes less time and the results are reliable. In this case study, you are going to explain how to organize and manipulate the data in order to reach the information we need to know. Median Mode and Mean As we said in our introduction, statistics are a way to summarize information. The data that we use have to be distributed and the main numerical property of a distribution is usually its central tendency. This notion represents some value around which distribution tends to centre. [...]

[...] The mean can also be useful to calculate the average of daily production in a factory. According to the analysis that we have done, we can say that sometimes, the mean isn't the best solution: - When data are qualitative (the mode is the most appropriate) - When the distribution is very skewed (the mean won't be affected by extreme values) . For instance, in a company 10 employees don't have the same wage per year 25 25 - 25 - 25 - 40 - 40 - 40 - 50 - 50 - 1000 The median: 40 The mean: 130 Of course, the good average would be the median one. [...]

[...] (The mean In series 1 and both means are really similar whereas in the series III, the mean is a bit higher. Why? Series 1 has a big spread between its extreme numbers: the lower number is 0,98 and the higher one is 21,54. Whereas in series II, the lower number is 8,02 and the higher is 11,69. So it is very strange that even if the data are completely different, the mean is quite the same. The 3 means are quite close but another key element, the variance, will show that this connection is due to other characteristics or variables of the distributions (The Variance First of all, it is important to explain what the variance means. [...]