Maman Abdurachman Djauhari

New York Stock Exchange performance: evidence from the forest of multidimensional minimum spanning trees

The standard practice of stock network analysis is still too far from reality. It involves Pearson correlation (PC) to quantify the similarity among stocks based on closing price only, and the minimum spanning tree (MST) to filter the economic information. In daily practice, stock is represented by its four prices: the opening, highest, lowest, and closing prices. To incorporate the information from these prices, vector correlation is used instead of PC. Furthermore, to get robust network topological properties, the MST is replaced by the forest of all possible MSTs. Its advantages is demonstrated in the analysis of 1515 stocks traded in the New York Stock Exchange from 2005–2014.

Dynamics of Correlation Structure in Stock Market

In this paper a correction factor for Jennrich’s statistic is introduced in order to be able not only to test the stability of correlation structure, but also to identify the time windows where the instability occurs. If Jennrich’s statistic is only to test the stability of correlation structure along predetermined non-overlapping time windows, the corrected statistic provides us with the history of correlation structure dynamics from time window to time window. A graphical representation will be provided to visualize that history. This information is necessary to make further analysis about, for example, the change of topological properties of minimal spanning tree. An example using NYSE data will illustrate its advantages.

Network topology of economic sectors

A lot of studies dealing with stock network analysis, where each individual stock is represented by a univariate time series of its closing price, have been published. In these studies, the similarity of two different stocks is quantified using a Pearson correlation coefficient on the logarithmic price returns. In this paper, we generalize the notion of similarity between univariate time series into multivariate time series which might be of different dimensions. This allows us to deal with economic sector network analysis, where the similarity between economic sectors is defined using Escoufiers vector correlation RV. To the best of our knowledge, there is no study dealing with this notion of economic sector similarity. Two examples of data from the New York stock exchange will be presented and discussed, and some important results will be highlighted.

Monitoring multivariate process variability when sub-group size is small

ABSTRACT The current practice of multivariate process variability monitoring, when sub-group size is small, has nothing to do with probability of false alarm (PFA). Consequently, the reliability of the existing control charts remains undetermined. In this article, we propose a control chart which is reliable, very sensitive to the change in variance for small or moderate correlation, and provides a root causes analysis of an out-of-control signal. To illustrate these advantages, an industrial example is presented and the results are compared with those issued from the existing methods.

A recommendation on PLUS highway development: A social network analysis approach

Road development is one of the important elements for the overall economic and social development in the world. A better interconnected road network is crucial to increase the economic activities, such as trade. Highway, as the backbone of a countrys transportation system, are required to overcome the demand for efficient road transportation. Besides, highway development is part of the government commitment towards the development of road network. In this paper, the network analysis of PLUS highway development was studied. The data used were based on the number of vehicles that entered and exited each toll plazas from Penang to Johor. The relationships between the toll plazas were studied by using minimum spanning trees and the overall centrality measures were used to determine the most influential toll plazas among the states. Based on the results, some recommendations are forwarded to the PLUS highway to improve the services and to increase the traffic efficiency.

Time series behaviour of the number of Air Asia passengers: A distributional approach

The common practice to time series analysis is by fitting a model and then further analysis is conducted on the residuals. However, if we know the distributional behavior of time series, the analyses in model identification, parameter estimation, and model checking are more straightforward. In this paper, we show that the number of Air Asia passengers can be represented as a geometric Brownian motion process. Therefore, instead of using the standard approach in model fitting, we use an appropriate transformation to come up with a stationary, normally distributed and even independent time series. An example in forecasting the number of Air Asia passengers will be given to illustrate the advantages of the method.

Asymptotic derivation of T* statistic

Monitoring the stability of correlation structure becomes an imperative subject in economic development and financial industry. It has been extensively implemented by various researchers to understand the behaviour of a sequence of correlation structures based on independent samples in certain time periods. The existing statistical test can only tackle problems involving p n due to the inversion of covariance matrix. Therefore, in this paper we introduced T* statistics, constructed based on upper-off-diagonal elements to overwhelm that difficulty. The limitation distribution has been investigated to make its application promising.

Sub-dominant ultrametric hierarchical structures of NYSE 100 stocks

In stock networks analysis sub-dominant ultrametric (SDU) is used to understand the economic hierarchical clustering among stocks. There are two different approaches to obtain SDU. First, which is usually used in stock networks analysis, is to construct a minimal spanning tree (MST) by using Kruskals algorithm or Prims and then derive the SDU. The second approach is to construct directly the SDU such as given by Johnsons algorithm. In this paper we examine the performance of the first approach based on Kruskals algorithm and the second approach based on Johnsons algorithm in terms of computational complexity and their running times by using NYSE 100 stocks daily prices data.

Logarithmic returns of electricity maximum loads during weekends and special days: Malaysia study case

Fitting a model in time series analysis can be very difficult when facing with random walk data. Logarithmic returns that usually used in financial modeling is one of the methods that can be used to form the suitable model for such time series. This paper deals with the advantages of logarithmic returns toward time series modeling of maximum electricity workloads in Malaysia. A case study on weekend days public holidays will be presented and discussed.

Network topology of Indonesian stock market

This paper proposes a new centrality measure called overall centrality measure that can be used to summarize the important information contained in social network. It is an optimal linear combination of the traditional measures where the optimality criterion is similar to that in principal component analysis. The advantages of this method will be illustrated by using stocks prices in Indonesian Stock Exchange where the relationship among stocks is viewed as social relationship. Some important results will be highlighted.