Bài tập thống kê trong kinh doanh – xác xuất thống kê eng

BÀI TẬP THỐNG KÊ TRONG KINH DOANH – XÁC XUẤT THỐNG KÊ Exercise 1 Graphing appropriate to assess the symmetry of the two methods above Statistical characteristics for two methods Find the 95% confidence interval for the average strength of the tire in each method Conducted testing to compare the average strength of the tires of the two methods above and conclude with α = 0.05 Exercise a Plotting the point spread to comment on the relationship possible between X and Y b Please estimate the relationship Simple linear regression filings and market quality products Conclusion? c Testing the relationship exists a linear correlation between the Y and X d Said coefficient R2 and explain its meaning e Please share if forecasts scale for product quality is 40, 50, 80 and 90 f According to you, whether only used as one variable X on Y has to explain enough If you can put more independent variables may be other factors? TABLE OF CONTENT TABLE OF CONTENT INTRODUCTION RESEARCH OF CONTENT Part Theoretical basics The core of management's decision making (decisions) .5 The components of decision making Part 2: Exercise .7 Exercise .7 a Draw the appropriate graph (BoxPlot) boxbot box to assess the symmetry of the data of the two methods mentioned above: .7 b As of the statistical characteristic of two methods and compare c Find the confidence interval 95% for the average strength of the tire in each method, see the average oscillation is d Please conduct testing to compare the average strength of tires in the methods and conclude with α = 0:05 e Make checks pair with α = 0.05 (assuming the data was obtained) Exercise 10 a Graphing scattered points to comment on the relationship possible between Y and X .10 b From the data obtained, by using regression analysis MegaStat/Correlation Regression/Regression Analysis of two parameters share of Y and the quality of the product X have the following results: 11 c To test the linear correlation between market share and product quality or not, it is assumed that there is no correlation with (β1 = 0) and we test the hypothesis pair follows: .11 d From the results of the regression analysis above we have R2 = 0896 This means that the change in market share Y is explained by 89.6% due to the variation in the quality of the product X 12 e Please share if forecasts scale for product quality is 40,50,80,90: regression analysis of the data of the two parameters Y and market share product quality X by megastat forecast market share in the case of 40 quality indicators, 50, 80 and 90 have been following results: .12 f According to you, is used as a variable X on Y has to explain enough If you can put more independent variables may be other factors? 13 LIST OF REFERENCES 14 INTRODUCTION The era we live in the "age of information", and for now, "the full utilization of information assets is a critical success factor, and now that has the ability to exploit maximum understanding will have a big advantage in the global business world "However, this scenario is still far away for enterprises in Vietnam, where the overwhelming information on the transmission medium, but the lack of tools to process information correctly The volume of data a company receives daily is huge because they come from so many parts of the company That's not to mention the external data transmission system Some enterprises not have a clear approach to system data into meaningful information and develop performance indicators (KPIs) from such information To obtain useful information for making management decisions, businesses need to go through four steps: collecting raw data, raw data classification, data analysis and screening information However, the first step is where most people were trapped How we can get accurate information and practical right time to make management decisions when initial data collection was spending too much time like that? This has slowed the decision making process, and thereby become a huge disadvantage to prevent the development of enterprises Every business needs to have accurate decisions to shape the future direction, and since then, development The only way to improve the quality of decisions is to improve information systems before, to provide practical information, consistency and application to decision makers In the next article, we will discuss more about the solution eliminates the three obstacles mentioned above, helps enterprises take the decision making to successful governance RESEARCH OF CONTENT Part Theoretical basics The core of management's decision making (decisions) The decision of the management, including: planning, choice of action plans, establishing organizational structure, division of responsibilities and powers, to compare the actual situation with the plan, the choice of test method investigation, take care of the present plan, organize and control for all levels of management and all aspects of the management process Decision of almost synonymous with management The management decisions are divided into two major groups: Decision overarching values are decisions on the final goal; decisions the fact that decisions relating to the implementation of the target (actual evaluation) The combination kinds of decisions that are considered to be the focus of management A management decision can only be considered valid if the element contains practical and feasible It was decisions "reasonable - objectivity" not decisions "reasonable - subjective" (optimal perfect) The components of decision making Decisions are made up of over four stages linked together: to collect and analyze information economy - society; design action plans to choose from; Choosing a viable option; verification assessment schemes have chosen to supplement perfect All activities within an organization can be divided into two categories with two kinds of decisions: activities took place many times require sequential decisions, and activities taking place for the first time need not follow policy decisions sequence It should try to raise the level of decision-making process to improve the control system effectively, while enhancing the air conditioning system, coordination of the organization Decisions of non-creative process, without precedent, but based on the knowledge and methods available to process, including the use of experience The division two types of decisions is relative Through a target system - the means to unify operations It is a combination of human and machine (electronic) to decision makers; can overcome the lack of knowledge and information in a communication network according to the twodimensional channels Choosing the right training methods or decentralization in decision-making powers to the authorities and clear Thanks to the automated determination of the order, the handling of issues related with the right training methods to be reasonable, to reduce the interference of intermediate to work at grassroots level However, that method can not be used in any situation; have to think about stimulus, making decisions can mobilize people trying to Organizational forms in the future is still the hierarchy consists of three levels: level of production and distribution, supply dominate decision making in sequence, control level during operation of the facility Need decentralized decision making, creating "the reasonable approval" for the autonomy of their subordinates The decisions taken with science, such decisions must be based on a plan The plan must set out the goals to be achieved The objective of the political society, there iêu item of business at the forefront it is profitable Measure, calculate profit or value of the different actions, we get the result or accounting In the calculation must take into account its revenues will be compared with the cost of removal Costs can be effective or ineffective Part 2: Exercise Exercise a Draw the appropriate graph (BoxPlot) boxbot box to assess the symmetry of the data of the two methods mentioned above: Through the data, to assess the symmetry of the data of the two methods above Can use the following graph: a Graph 1: Boxplot by the method Graph 1: Boxplot by the method Reviews: Then look at the two graphs on the data we see method is relatively symmetrical because due to the length of the equivalent box Method misses to the left b As of the statistical characteristic of two methods and compare Descriptive statistics Count Mean Sample variance Sample standard deviation Minimum Maximum Range 1st quartile Median Method 40 2,742.58 1,087.94 32.98 2685 2799 114 2,714.50 2,740.50 Method 40 2,729.35 1,468.34 38.32 2660 2789 129 2,689.00 2,738.50 3rd quartile 2,774.50 2,757.75 Interquartile range 60.00 68.75 Mode 2,755.00 2,676.00 Low extremes 0 Low outliers 0 High outliers 0 High extremes 0 Commented: We saw both methods are 40 observations The average value equivalent (2742 & 2729) Comparing the median well worth the equivalent 2,740.50 2,738.50 Comparison of dispersion method has a larger dispersion by the standard deviation and interquartile range are greater than method Both methods are no outliers c Find the confidence interval 95% for the average strength of the tire in each method, see the average oscillation is Descriptive statistics Method Method Count 40 40 Confidence interval 95.% lower 2,732.03 2,717.10 Confidence interval 95.% upper 2,753.12 2,741.60 Half-width 10.55 12.25 Conclusion: With the reliability 95% of the average strength for method was from 2,732 to 2,753 As for method is 2,717 to 2,741 d Please conduct testing to compare the average strength of tires in the methods and conclude with α = 0:05 We have; Ho ; ℳ1 = ℳ2 H1 ; ℳ1# ℳ2 In which ℳ1 : Strength of tires in Method ℳ2 : Strength of tires in Method Hypothesis Test: Independent Groups (t-test, pooled variance) PP1 2,742.58 32.98 40 PP2 2,729.35 Mean 38.32 std dev 40 N 78 13.225 1,278.139 35.751 7.994 1.65 1021 Df difference (PP1 – PP2) pooled variance pooled std dev standard error of difference hypothesized difference T p-value (two-tailed) Conclusion: P-value = 0.1021 is greater than α = 0.05 should not reject H0 It is concluded tire endurance average of the two methods are the same (as the person in question number two) e Make checks pair with α = 0.05 (assuming the data was obtained) Then, we saw a few broken tire on the road During the investigation, on the assumption was made A record sheet for the test was conducted with 40 pairs of randomly selected samples other Each tires in pairs are produced according to two different methods (in phase with the improvement of methods), while all other stages in the production process is the same It is suggested that the test is more appropriate pair If this is considered experimental in pairs, we use the data pairs in which each pair of Method and Methed included Using software datasets pair Hypothesis Test: Paired Observations 0.000 2,742.575 2,729.350 13.225 48.480 7.665 40 39 1.73 0924 hypothesized value mean PP1 mean PP2 mean difference (PP1 - PP2) std dev std error n df t p-value (two-tailed) With a p-value = 0.0924 is greater than α = 5% should also not rejected H So if using pairs of data, the average strength of 02 tires method is considered to be the same Exercise a Graphing scattered points to comment on the relationship possible between Y and X CHART POINT SPREAD MARKET SHARE IN QUALITY 10 b From the data obtained, by using regression analysis MegaStat/Correlation Regression/Regression Analysis of two parameters share of Y and the quality of the product X have the following results: Regression analysis was used to predict the value of the dependent variable based on the value of at least one independent variable Explain the impact of the change in the independent variable on the dependent variable The relationship between the two variables Y and X are described by a linear function Regression Analysis r² r Std Error ANOVA table Source Regression Residual Total 0.896 0.946 1.301 SS 217.6803 25.3786 243.0588 n k Dep Var df 15 16 MS 217.6803 1.6919 17 Y F 128.66 Regression output variables Intercept X p-value 9.30E-09 confidence interval 95% coefficients -2.7054 0.1784 std error 1.0593 0.0157 t (df=15) -2.554 11.343 p-value 0220 9.30E-09 95% lower -4.9632 0.1449 upper -0.4476 0.2119 So, the two regression parameters are written as follows Y = β0 + β1X = -2.7054 + 0.1784X Blocking coefficient β0= -2.7054 The slope β1 = 0.1784 Conclusion: Do β1> so Y and X variables in the same direction so as quality improves, the market share will increase If quality changes one point they changed 0.1784% market share c To test the linear correlation between market share and product quality or not, it is assumed that there is no correlation with (β1 = 0) and we test the hypothesis pair follows: 11 H0: β1 = There is no linear relationship H1: β1 # There is a linear relationship From the results of the regression analysis above we see p value = 9.30E-09-