臺灣博碩士論文加值系統

在本研究中，我們想了解當我們在面對截斷變數或Tobit-like變數的模組時。是否能找出一個在資料不同分配下，較為強健的模組；因此，本研究選擇了一般面對截斷變數或Tobit-like變數的模組時，較為常用的對數常態模組以及Cragg’s模組來作為兩相較模組；並利用模擬方法，得出不同分配的資料，並採用Vuong所提出的LR檢定來作比較。發現在兩相較模組下，Cragg’s模組會比對數常態模組來的穩定及強健許多。所以，我們採用Cragg’s模組拿來與傳統Tobit模組來作比較，利用R&D的實證資料，經由KLIC法則，我們發現在此實證資料當中，我們可以定下一個結論，就是當我們在處理截斷變數或Tobit-like變數的模組時，可以優先考量使用較穩定的Cragg’s模組來作分析。

In this study, we want to identify the robust model in different distribution data, especially in facing censored variables or Tobit —like variables. In order to verify this thought, we choose two competitive models: lognormal model and Cragg’s model for examination. With two different kinds distribution simulated data, we use Voung’s model selection tests for two competitive hurdle, or two-tier models. In these simulated data, we find out that Cragg’s model will be more robust than lognormal model. So, we take it to compare with the traditional Tobit model, for another suggestion when researching the R&D expenditure. After testing through KLIC rule with real data, we can find that the Cragg’s model is more suitable than Tobit’s model in this real data set.

Contents
Contents i
Table catalogs ii
I. Abstract 1
II. Introduction 2
2.1 Research Objectives 5
2.2 Research Motivation 6
III. Literature Review 6
IV. Research Methodology 10
4.1 Model selection for two-part model with simulated data 10
4.1.1 Simulation Data 10
4.1.2 Two competitive model 12
4.1.3 Kullback-Leibler Information Criterion 14
4.1.4 Parameter Settings 16
4.2 Real Data used in R&D 16
4.2.1 Investigated Period 16
4.2.2 Data source 16
4.3 Estimate Model and Model Specification 17
4.3.1 Model Hypothesis 17
4.3.2 Model selection between Tobit’s and Cragg’s model 17
V. Empirical Results 19
5.1 Model selection result with simulation data 19
5.2 Real data result 20
VI. Conclusions and suggestions 20

English
1. Cox, D. R. (1962): "Further Results on Tests of Separate families of Hypotheses," Journal of Royal Statistical Society, Series B, 24, 406-424.
2.Cragg, J. (1971), "Some Statistical Models for Limited Dependent Variables with Application to the Demand for Durable Goods," Econometrica 39,829-844.
3.Fin, T. and P. Schmidt (1984), "A Test of the Tobit Specification Against an
Alternative Suggested by Cragg," Review of Economics and Statistics 66, 174-177.
4.Heckman, J. (1976). "The Common Structure of Statistical Models of Truncation,
Sample Selection and Limited Dependent Variables and a Simple Estimator for
Such Models." Annals of Economic and Social Measurement 5:475-92.
5.Kullback, S., and R. A. Leibler (1951): "On Information and Sufficiency," Annals of Mathematical Statistics, 22, 79-86.
6.Margrita, G. (2001), “A note about model selection and tests for non-nested contingent valuation models”. Economics Letters 74.363-370
7.Sin, C.Y., White H.,(1996) “Information criteria for selecting possibly misspecified parametric models. Journal of Econometrics 71,207-225
8.Vella, R (1993), "A Simple Estimator for Simultaneous Models with Censored
Endogenous Variables," International Economic Review 34, 441 -457.
9.Vuong’Q (1989), "Likelihood Ration Tests for Model Selection and Nonested Hypotheses," Econometrica 57, 307-333.
10.White, H. (1982): "Maximum Likelihood Estimation of Misspecified Models " Econometrica, 50, 1-25
11.J White, H. (1982). Regularity Conditions for Cox's Test of Non-Nested Hypotheses "Journal of Econometrics, 19, 301-318.
12.White, H. (1982), "Instrumental Variables Regression with Independent Obervations " Econometrica 50, 483-500.
13.Willis, R., and S. Rosen (1979). "Education and Self-Selection." Journal of Political
Economy 87(5, Part 2):507-36.
中文:
1.方至民等(2000)，高階經營團隊特質與財務能力對企業研發支出影響之研究--以臺灣上市電子產業公司為例，科技管理學刊，5:3 頁29-46
2.歐進士(1998)，我國企業研究發展與經營績效關聯之實證研究，中山管理評論，夏，頁357-385
3.鄭嘉珮(1990)，廠商研究發展支出的決定因素─國資與外資廠商之實證，政治大學國際貿易研究所
4.徐子光(1987)，產業之研究發展及其影響因素分析，台灣經濟研究院，卷112，
頁38-44
5.張恩浩(1991)，研究發展之影響因素及其與績效關連，台灣大學商學研究所論文
6.黃雅苓(1998)，研究發展支出與經營績效關係及其費用化之控討-以台灣上市公司之電子業與非電子業為例，國立政治大學會計研究所碩士論文