KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Access type: UnknownAuthor: search.filters.author.Narabin, Santit

Search Results

Now showing 1 - 5 of 5
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    A Cluster Analysis of Mutual Funds Data
    (2018-11-26)
    Narabin, Santit
    ;
    Boongasame, Laor
    The factors of clustering mutual fund (such as Net Asset Value (NAV)) do not direct to both return and risk of mutual funds which they are important factors for investors. This research helps an investor can estimate profit and loss rate of the mutual fund in his/her portfolio by using the net asset value change ratios (NAVCR). Then, both the NAVCR and value of each mutual fund will be used for clustering. For building a portfolio, the mutual funds could be selected from the diversified groups in order to reduce risk. The mutual fund data at different times from the set for the fiscal year 2010 - 2017 are used. The results of our analysis show that our models offer significantly better performance than the portfolio management model derived from the random portfolio management.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Selecting students to a dormitory using AHP
    (2016-11-18)
    Narabin, Santit
    ;
    Boonjing, Veera
    A dormitory is an accommodation service of a university for the students especially who live far away from the educational institutions. In our country, with perfectly low price, a dormitory is needed for number of students. Each year student applicants are selected by a committee. It can be happened that a group of students may not be selected by a committee member without a reasonable reason. This fact can bring undesirable biases in educational area. Therefore, using rational and acceptable criteria for selecting students to stay in a dorm are needed. In this article, we propose to apply Analytical Hierarchy Process (AHP) as a Decision Support System (DSS) to select undergraduate students to a dormitory of a university. We provide 4 main criteria and 16 sub-criteria. Based on the method, an eigenvector is calculated and the Consistency Ratio (CR) can be determined from the eigenvector. By iteratively following the process of AHP, we then can have appropriate weight numbers that can be used for selecting proper students to the dormitories.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    A new coalition formation based on reservation prices and locations of buyers
    (2013-01-01)
    Narabin, Santit
    ;
    Boonjing, Veera
    ;
    Boongasame, Laor
    This paper defines a coalition as a cluster of buyers in dimensional space of reservation price and location. It defines a new buyer utility accounting for discount and traveling costs. A new coalition with nonnegative utility is proposed to maximize a number of successful buyers. This nonnegative utility coalition also assures that a winner does not have to pay more to subsidize others. Simulation results confirm that performance of the new scheme is significantly better than of a random one and is as good as a reservation price based coalition. © 2013 S. Narabin et al.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    A location-price-based buyer coalition
    (2012-12-01)
    Narabin, Santit
    ;
    Boonjing, Veera
    A buyer coalition is a group of buyers who join together to negotiate with sellers to purchase items for a larger discount. In this article, a novel buyer coalition scheme, called the "GroupSimilarBuyer Scheme, " is introduced. The new idea focuses on buyer coalition formation based on similarity of buyers. The mechanism of approach begins to prepare data for driving the next step. Then, a k-mean clustering algorithm is used to form the coalition of buyers which are very similar in both reservation price and location of buyer. In addition, the utility of each buyer coalition is calculated. Based on the simulation result, the GroupSimilarBuyer showed that the average standard deviation of the coalition was nearly the optimum result. © 2012 IEEE.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    An integer partition based algorithm for coalition structure generation
    (2007-12-01)
    Boonjing, Veera
    ;
    Narabin, Santit
    This paper proposes a new algorithm to generate a minimal search space of the problem of coalition structure generation using a new optimal integer partition. The new partition includes only partitions giving optimal coalition structures. These partitions are those containing 1 at most one element. Our new algorithm generates optimal partition structures in two steps. The first step, we use a modified version of ZS1 algorithm to generate the optimal integer partition of input integer. We then use the partition to generate optimal coalition structures in the second step. © 2007 IEEE.