Estimating the Number of Clusters in a Data Set Via the Gap Statistic

Tibshirani, Robert; Walther, Guenther; Hastie, Trevor

doi:10.1111/1467-9868.00293

Cited by 4,842 publications

(3,617 citation statements)

References 18 publications

Supporting

Mentioning

3,554

Contrasting

Unclassified

Order By: Relevance

“…In our application, the same result obtains if we use the gap statistic developed by Tibshirani, Walther and Hastie (2001). 11 "If they want to take us out from every one of the (congressional) Committees, let them do it; we have the streets of the people" -our translation.…”

Section: The Conditional Effects Of Constitutions On Policysupporting

confidence: 53%

Are We All Playing the Same Game? The Economic Effects of Constitutions Depend on the Degree of Institutionalization

2013

View full text Add to dashboard Cite

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. The understanding of the economic effect of formal institutional rules has progressed substantially in recent decades. These formal analyses have tended to take for granted that institutional arenas such as Congress are the places where decision-making takes place. That is a good approximation in some cases (such as many developed countries today) but not in others. If countries differ in how institutionalized their policymaking is, it is possible that the impact of formal political rules on policy outcomes might depend on that. This paper explores that hypothesis and finds that some important claims regarding the impact of constitutions on policy outcomes do not hold for countries in which institutionalization is low. The findings suggest the need to develop a broader class of policymaking models in which the degree to which decision-making follows "the rules" is also endogenized. Terms of use: Documents inJEL Codes: D78, D72, D73, H60, H62, H20

show abstract

Section: The Conditional Effects Of Constitutions On Policysupporting

confidence: 53%

Are We All Playing the Same Game? The Economic Effects of Constitutions Depend on the Degree of Institutionalization

2013

View full text Add to dashboard Cite

show abstract

“…The maximum number of clusters was defined to be equal to the number of variables. Different criteria exist to identify the optimal number of clusters in a given data set (Tibshirani et al, 2001;Yan, 2005) but in this study, the aim was to identify cluster patterns at different stages of clustering instead of identifying one optimal number of clusters. Relevant cluster stages were defined as stages resulting in a high increase in variance explained compared to the neighbouring splits (local maxima).…”

Section: Methodsmentioning

confidence: 99%

Investigating integument alterations in cubicle housed dairy cows: which types and locations can be combined?

Brenninkmeyer

Dippel

Brinkmann

et al. 2016

animal

View full text Add to dashboard Cite

In this study, a data set of 2922 lactating dairy cows in a sample of 64 conventional and organic dairy farms with Holstein Friesian cows in Germany and 31 conventional dairy farms with the dual purpose breed Fleckvieh in Austria was used to screen for correlations between the occurrences of different integument alterations. All cows were housed in cubicle systems. Alterations were classified as hairless areas (H), scabs or wounds (W) or swellings (S) and assessed at 15 locations of the cows' body. Highest median farm prevalences were found at the joints of the legs, which are already commonly included in studies on integumentary alterations: median farm prevalence was 83% for S and 48% for H at the carpal joints, followed by H (38%) and S (20%) at the lateral tarsal joints and H at the lateral calcanei (20%). Additional body parts with notable median prevalences for H were the hip bones (13%), pin bones (12%) and sacrum (11%). Three cluster models, with 2, 5 and 14 clusters, were built by hierarchical clustering of prevalences of the 30 most relevant alteration location combinations. Clustering revealed that location overruled type of lesion in most cases. Occasionally, clusters represented body segments significantly distant from each other, for example the carpal joints and lateral and dorsal calcanei. However, some neighbouring areas such as the medial and lateral hock area should be analysed separately from each other for causal analysis as they formed distinct clusters.

show abstract

“…Prior literature (e.g., Smyth 2000; Still and Bielek 2004;Tibshirani et al 2001) recommends an iterative process to determine the optimal number of clusters. Consistent with this idea, we experimentally varied the number of clusters and repeated the three steps in Figure 2 until the error rate of the decision tree model in Step 3 was minimized.…”

Section: Classification Modelmentioning

confidence: 99%

The Association Between the Disclosure and the Realization of Information Security Risk Factors

Wang

Kannan

Ulmer

2013

Information Systems Research

150

View full text Add to dashboard Cite

Firms often disclose information security risk factors in public filings such as 10-K reports.The internal information associated with disclosures may be positive or negative. In this paper, we are interested in evaluating how the nature of security risk factors disclosed, which is believed to represent the internal information regarding information security, is associated with future breach announcements. For this purpose, we build a decision tree model, which classifies the occurrence of future security breaches based on the textual contents of the disclosed security risk factors. The model is able to accurately associate disclosure characteristics with breach announcements about 77% of the time. We further explore the contents of the security risk factors using text mining techniques to provide a richer interpretation of the results. The results show that the security risk factors with action-oriented terms and phrases are less likely to be related to future incidents. We also conduct a cross-sectional analysis to study how the market interprets the nature of information security risk factors in annual reports at different time points.We find that the market reaction following the security breach announcement is different depending on the nature of disclosure. Thus, our paper contributes to the literature in information security and sheds light on how market participants can better interpret security risk factors disclosed in financial reports at the time when financial reports are released.

show abstract

Estimating the Number of Clusters in a Data Set Via the Gap Statistic

Cited by 4,842 publications

References 18 publications

Are We All Playing the Same Game? The Economic Effects of Constitutions Depend on the Degree of Institutionalization

Are We All Playing the Same Game? The Economic Effects of Constitutions Depend on the Degree of Institutionalization

Investigating integument alterations in cubicle housed dairy cows: which types and locations can be combined?

The Association Between the Disclosure and the Realization of Information Security Risk Factors

Contact Info

Product

Resources

About