The Experts below are selected from a list of 1347 Experts worldwide ranked by ideXlab platform

Elizabeth G. Hill - One of the best experts on this subject based on the ideXlab platform.

  • Ordinal Logic Regression
    Computational Statistics & Data Analysis, 2015
    Co-Authors: Bethany J. Wolf, Elizabeth H. Slate, Elizabeth G. Hill
    Abstract:

    In medicine, it is often useful to stratify patients according to disease risk, severity, or response to therapy. Since many diseases arise from complex gene-gene and gene-environment interactions, patient strata may be defined by combinations of genetic and environmental factors. Traditional statistical methods require specifying interactions a priori making it difficult to identify high order interactions. Alternatively, machine learning methods can model complex interactions, however these models are often difficult to interpret in a clinical setting. Logic regression (LR) enables modeling a binary outcome using Logical combinations of binary predictors yielding easily interpretable models. However LR, as currently available, cannot model Ordinal responses. This paper extends LR to model an Ordinal response and the resulting method is called Ordinal Logic Regression (OLR). Several simulations comparing OLR and Classification and Regression Trees (CART) demonstrate that OLR is superior to CART for identifying variable interactions associated with an Ordinal response. OLR is applied to data from a study to determine associations between genetic and health factors with severity of adult periodontitis. Ordinal Logic Regression is publicly available on CRAN in the OrdLogReg package, http://cran.r-project.org/.

Gisèle Umbhauer - One of the best experts on this subject based on the ideXlab platform.

  • Second-Price All-Pay Auctions and Best-Reply Matching Equilibria
    International Game Theory Review, 2019
    Co-Authors: Gisèle Umbhauer
    Abstract:

    The paper studies second-price all-pay auctions — wars of attrition — in a new way, based on classroom experiments and Kosfeld et al.’s best-reply matching (BRM) equilibrium. Two players fight over a prize of value [Formula: see text], and submit bids not exceeding a budget [Formula: see text]; both pay the lowest bid and the prize goes to the highest bidder. The behavior probability distributions in the classroom experiments are strikingly different from the mixed Nash equilibrium (NE). They fit with BRM and generalized best-reply matching (GBRM), an Ordinal Logic according to which, if bid A is the best response to bid B, then A is played as often as B. The paper goes into the GBRM Logic, highlights the role of focal values and discusses the high or low payoffs this Logic can lead to.

  • Second price all-pay auctions, how much money do players get or lose?
    2017
    Co-Authors: Gisèle Umbhauer
    Abstract:

    The paper studies second price all-pay auctions - wars of attrition - in a new way, based on class room experiments and Kosfeld, Droste and Voorneveld’s (2002) best reply matching equilibrium. Two players fight over a prize of value V, have a budget M, submit bids lower or equal to M; both pay the lowest bid and the prize goes to the highest bidder. The behaviour probability distributions in the class room experiments are strikingly different from the mixed Nash equilibrium. They fit with best reply matching or generalized best reply matching, an Ordinal Logic according to which, if bid A is the best response to bid B, and if B is played with probability p, then A is also played with probability p. In the mixed Nash equilibrium, the expected payoff is never negative and close to 0. In the best reply and generalized best reply matching equilibria, players may lose money, up to 1/12th of the budget when M is large in comparison to V, but they can also get a lot of money, especially if V is large. The study leads to examine possible bifurcations in the bidding behaviour and gives some insights into how to regulate games to avoid pathoLogical gambling with a huge waste of money.

G Hillelizabeth - One of the best experts on this subject based on the ideXlab platform.

  • Ordinal Logic Regression
    Computational Statistics & Data Analysis, 2015
    Co-Authors: J Wolfbethany, H Slateelizabeth, G Hillelizabeth
    Abstract:

    In medicine, it is often useful to stratify patients according to disease risk, severity, or response to therapy. Since many diseases arise from complex gene-gene and gene-environment interactions,...

Bethany J. Wolf - One of the best experts on this subject based on the ideXlab platform.

  • Ordinal Logic Regression
    Computational Statistics & Data Analysis, 2015
    Co-Authors: Bethany J. Wolf, Elizabeth H. Slate, Elizabeth G. Hill
    Abstract:

    In medicine, it is often useful to stratify patients according to disease risk, severity, or response to therapy. Since many diseases arise from complex gene-gene and gene-environment interactions, patient strata may be defined by combinations of genetic and environmental factors. Traditional statistical methods require specifying interactions a priori making it difficult to identify high order interactions. Alternatively, machine learning methods can model complex interactions, however these models are often difficult to interpret in a clinical setting. Logic regression (LR) enables modeling a binary outcome using Logical combinations of binary predictors yielding easily interpretable models. However LR, as currently available, cannot model Ordinal responses. This paper extends LR to model an Ordinal response and the resulting method is called Ordinal Logic Regression (OLR). Several simulations comparing OLR and Classification and Regression Trees (CART) demonstrate that OLR is superior to CART for identifying variable interactions associated with an Ordinal response. OLR is applied to data from a study to determine associations between genetic and health factors with severity of adult periodontitis. Ordinal Logic Regression is publicly available on CRAN in the OrdLogReg package, http://cran.r-project.org/.

Nnaemeka Ikpeze - One of the best experts on this subject based on the ideXlab platform.

  • Assessing Infrastructural Development in Ogoniland in the Niger Delta Region of Nigeria: A Participatory Development Approach
    2015
    Co-Authors: Makuachukwu Gabriel Ojide, Emmanuel O. Nwosu, Josiah O. Aramide, Nnaemeka Ikpeze
    Abstract:

    Availability and access to basic infrastructure in rural areas across Nigeria has been identified by many as a crucial component for national economic development. The impacts of infrastructural facilities on quality of life and overall development cannot be underestimated. In light of the above, this paper assesses infrastructural development in Ogoniland through a participatory development approach. Using a multistage sampling method, four hundred households were surveyed in the community. The Ordinal Logic regression model indicates that infrastructural development in Ogoni community has been influenced by government interventions and oil companies through their corporate social responsibility programmes.