IID three discrete selection models MNL, MNP, MIXED LOGIT.

http://math.stackexchange.com/questions/466927/independent-identically-distributed-iid-random-variables

independent, identically distributed (IID) random variables

up vote5down votefavorite 7

I am having trouble understanding IID random variables. I've tried readinghttp://scipp.ucsc.edu/~haber/ph116C/iid.pdf,http://www.math.ntu.edu.tw/~hchen/teaching/StatInference/notes/lecture32.pdf, andhttp://www-inst.eecs.berkeley.edu/%7Ecs70/sp13/notes/n17.sp13.pdf but I don't get it. Would someone explain in simple terms what IID random variables are and give me an example? probability probability-theory random-variables independence shareciteimprove this question edited Nov 12 '15 at 9:25 BCLC 4,3171933 asked Aug 13 '13 at 21:35 Frank Epps 2771217

say you are sampling from a known distribution with replacement. Every time you, say, draw a sample, this is a random variable. Drawn samples are independent of each other, and the distribution never changes. Thus, IID random variables – Eleven-Eleven Aug 13 '13 at 21:41 add a comment

2 Answers

activeoldestvotes

up vote12down voteaccepted

Im sure you know that iid means independent, identically distributed. I think the most prominent example is a coin toss repeated several times. If X1,X2,…$X_1,X_2,\dots$ designate the result of the 1st, 2nd, and so on toss (where Xi=1$X_i=1$ means that in the i-th toss you have got head and Xi=0$X_i=0$ tail), you have that X1,X2,…$X_1,X_2,\dots$ are iid. They are independent since every time you flip a coin, the previous result doesn't influence your current result. Edit: there is a mathematical definition of independence, but I don't think it is necessary at the moment. They are identically distributed, since every time you flip a coin, the chances of getting head (or tail) are identical, no matter if its the 1st or the 100th toss (probability distribution is identical over time). If the coin is "fair" the chances are 0,5 for each event (getting head or tail). Does that help? shareciteimprove this answer edited Aug 13 '13 at 23:41 answered Aug 13 '13 at 21:41 Dima McGreen 420312

add a comment

up vote0down vote

.So basically you will consider events where the outcome in one case will not depend on the outcome of the other cases .It is called identical because in every case u consider the possible outcomes will be same as the previous event .Some one has suggested yes tossing of coin is a good example .I will try to be a statistician here .You should go through few statistical distributions like normal ,gamma etc and then see additive property .There while proving the additive property u will consider independent events initially and then prove that the addition of all the independent variates also follows the respective distribution using the MGF of that particular distribution and then you will extend your property to see what if the variates are made similar .Hope you got your answer shareciteimprove this answer answered Mar 11 at 14:26 Tejas Suresh 11

Welcome to Math.SE. This question has a well-accepted answer. You have provided nothing new. And what you have provided does not exactly answer the question. Please refrain from opening up old questions which have good answers unless you have something significant to contribute. THere are many new questions begging for answers. – Shailesh Mar 11 at 14:46

1] 陈锟、王晓红、朱敏（2006），“Logit模型在市场营销定量研究中的应用” [J]，
《商业研究》24：196-199 [2] 董加伟，王方华（2004），“基于离散选择模型DCM的产品差异市场分析”[J]，《运
筹与管理》，2：79-84
[3] 郭永新、王高、齐二石（2007），“品牌、价格和促销对市场份额影响的模型研究”
[J]，《管理科学学报》， 2：p60-65 [4] 李德春、肖岳峰（2007），“基于多项式Logit模型的快速消费品促销效果评估” [J]，
《桂林电子科技大学学报》2：74-77 [5] 聂冲、贾生华（2005），“离散选择模型的基本原理及其发展演进评价” [J]，《数
量经济技术经济研究》11：151-159 [6] 王树盛，黄卫，陆振波（2006），“Mixed Logit模型及其在交通方式分担中的应用
研究”[J]，《公路交通科技》，2006，5：88-91 [7] 王树盛，黄卫，葛华（2006），“Mixed Logit模型随机参数分布形式选择探讨”[J]，
《公路交通科技》，8：122-125 [8] 吴中，毛荣昌(2004)，“MNL模型在交通方式划分中的新发展”[J]，《交通标准化》，
7：40-42 [9] 袁诚（2003），“离散选择模型在产品差别研究中的应用”[J]，《统计研究》， 1：
52-56 [10] 张群、曹丽、李纯青（2005），“离散选择模型的比较及其在零售业中的应用” [J]，
《西安工业学院学报》3：293-298 [11] 邹德强、王高、赵平、王燕（2007），“功能性价值和象征性价值对品牌忠诚的影
响:性别差异和品牌差异的调节作用” [J]，《南开管理评论》，3：4-13 [12] Ben-Akiva, M.(1972), The Structure of Travel Demand Models[D], PHD Thesis
[13] Ben-Akiva, Moshe, and Steven R. Lerman(1985), Discrete Choice Analysis: Theory
and Application to Travel Demand, Cambridge, MA: The MIT Press
[14] Bhat, C.(1998), Accommodating Variations in Responsiveness to Level-of Service
Variables in Travel Mode Choice Models[J], Transportation Research A 32:p495-510 [15] Bhat, C.(2001), Incorporating Observed and Unoberved Heterogeneity in Urban Work
Mode Choice Modeling[J], Transportation Research A 34:p228-238
[16] Buzzell, R. D. and B. T. Gale（1987）, The PIMS Principles: Linking Strategy to
Performance [M], New York, NY: The Free Press.
[17] Chintagunta, Pradeep K（1992）, Estimating A Mutinomial Probit Model of Brand
Choice Using the Method of Simulated Moments [J], Marketing Science, Fall, Vol. 11 Issue 4, p386, 22p  [18] Erdem, Tülin（1996）, A Dynamic Analysis of Market Structure Based on Panel Data[J],
Marketing Science , Vol. 15 Issue 4, p359, 20p
[19] Green, Paul E., Frank J. Carmone, and David P. Wachspress (1977), On the Analysis of
Qualitative Data in Marketing Research, Journal of Marketing Research, Vol. 18 (February), p52-9
[20] McFadden, Daniel(1974), Conditional logit analysis of qualitative choice behavior[M],
Frontiers in Econometrics, New York: Academic Press, 105-142
[21] Russell, Gary J. and Ann Petersen (2000), Analysis of Cross Category Dependence in
Market Basket Selection, Journal of Retailing, 76 (Fall), p367-392









 18
[22] Train, K. , D. McFadden, and A. Goett (1987), “Consumer attitudes and voluntary rate
schedules for public utilities”[J],Review of Economics and Statistics, LXIX(3):383-391 [23] Train(1999), Mixed Logit Models for Recreation Demand[M], in J. Herriges and C.
Kling, eds., Valuing Recreation and the Enviroment, Edward Elgar, Northampton [24] Train（2002）, Discrete Choice Methods with Simulation [M], Cambridge University
Press, p15-16
[25] Zsolt SÁndor, Michel Wedel (2005)，Heterogeneous Conjoint Choice Designs.
Journal of Marketing Research 42:2, 210-218

The Application of Discrete Choice Model on Retailing Research –Based on Comparation between MNL、MNP and Mixed Logit

Liu Yong, Wang Gao

Abstract  Discrete Choice model has been used widely in many research fields such as traffic, reas estate, tele-communication, energy and marketing in the world, however the model in marketing research is still in its infancy in China. This paper applies three discrete selection models MNL, MNP, MIXED LOGIT. It focuses on the impact of marketing devices on customers' brand selction and market share in Sanitary Pad market. The author found that the brand and various other promotional factors are playing positive impact when customers purchasing decisions are made. Particularly, some promotional means, such as Gift-promotion, which can bring customers some visible advantages, effect the greatest.