Registered Data

[CT087]

[02487] Asset Forecasting Using Geometric Brownian Motion and Variance Gamma Models

  • Session Date & Time : 3E (Aug.23, 17:40-19:20)
  • Type : Contributed Talk
  • Abstract : The basic assumption in the Black-Scholes-Merton model is log returns assets normally distributed. In reality, asset price movements are so fluctuating that the data is not normally distributed. This paper proposes a way to forecast using the variance gamma (VG) model. The VG model has three parameters to control volatility, skewness, and kurtosis. We compare results with the geometric Brownian motion (GBM) model. The accuracy of the model used the mean absolute percentage error (MAPE).
  • Classification : 62P20
  • Author(s) :
    • Abdul Hoyyi (Gadjah Mada University)
    • Abdurakhman Abdurakhman (Gadjah Mada University)
    • Dedi Rosadi (Gadjah Mada University)

[02572] LDA Hyper-parameters Regulation for Qualitative Studies in Management

  • Session Date & Time : 3E (Aug.23, 17:40-19:20)
  • Type : Contributed Talk
  • Abstract : The suitability of the number of topics in management qualitative research is a complex problem. Topic models on textual data often require judgment expertise in determining suitability for management problems. One of the things to note is the determination of hyperparameter values. We have designed various hyper-parameter value conditions as an experiment and calculated the document match probability based on the topic grouping results. Different hyperparameter values indicate different levels of probability.
  • Classification : 62P25, 68W99, 91C20
  • Author(s) :
    • Evita Purnaningrum (Mathematics Department, Gadjah Mada University)
    • Abdurakhman Abdurakhman (Mathematics Department, Gadjah Mada University)
    • Nanang Susyanto (Mathematics Department, Gadjah Mada University)

[01068] Reverse engineering controller area network using the Pearson correlation coefficient

  • Session Date & Time : 3E (Aug.23, 17:40-19:20)
  • Type : Industrial Contributed Talk
  • Abstract : Controller Area Network \(CAN\) is a communication bus widely adopted in road vehicles. However, car manufacturers adopt proprietary CAN message sets, complicating message decoding by third-party applications, e.g., for indirectly detecting adverse road-weather conditions. This talk presents an algorithm for reverse engineering CAN messages based on the Pearson correlation coefficient between CAN messages from an annotated and interpolated dataset. The proposed algorithm was experimentally validated with data collected from different vehicle brands.
  • Classification : 62P30, 68P20, 62-04
  • Author(s) :
    • David Rocha (Instituto de Telecomunicacoes, Universidade de Aveiro)
    • João Almeida (Instituto de Telecomunicações, Universidade de Aveiro)
    • José Fonseca (Instituto de Telecomunicações, Universidade de Aveiro)
    • Joaquim Ferreira (Instituto de Telecomunicações, Universidade de Aveiro)

[00343] A wavelet-based methodology to compare the impact of COVID-19 pandemic versus Russia-Ukraine conflict on crude oil sector and its interconnectedness with other energy and non-energy markets

  • Session Date & Time : 3E (Aug.23, 17:40-19:20)
  • Type : Contributed Talk
  • Abstract : We address the important question of whether the COVID-19 pandemic and the Russia-Ukraine conflict have the same economic impact. Our study focuses on a comparative analysis of the crude oil markets with other energy and non-energy markets in terms of market efficiency and interconnectedness during the two crises. We deploy a novel methodology that utilizes the wavelet decomposition of time series, which facilitates capturing of information in both time and frequency domain, to evaluate the market efficiency of these markets. Subsequently, the wavelet coherence, along with a vector-valued GARCH model applied to the wavelet details, helps us in quantifying the interconnectedness and spillover dynamics between different markets. Our analysis suggests that the energy sector is impacted more than the non-energy sector in times of both crises, however, the nature of the impact is different for different energy markets. Brent crude oil suffers more during the Russia-Ukraine war than during the pandemic, while natural gas suffers more during the pandemic than the war and WTI suffers equally during both crises. We also report increased interconnectedness between markets during the pandemic and the war, however, the degree of comovement varies from one time scale to another. This information would help investors to choose a safe market and plan their portfolio accordingly in a crisis period.
  • Classification : 62Pxx, Financial markets, stochastic models in economics, energy economics
  • Author(s) :
    • Archi Roy (Doctoral student (Indian Institute of Science, Education and Research, Pune))
    • Anchal Soni (Doctoral student (Indian Institute of Management, Banglaore))
    • Soudeep Deb (Assistant professor (Indian Institute of Management, Bangalore))

[00706] Peak and Short-term Electricity demand using Kalman Filtered Monte-Carlo Method

  • Session Date & Time : 3E (Aug.23, 17:40-19:20)
  • Type : Contributed Talk
  • Abstract : The study seeks to predict the next day’s peak and energy demand in Ghana. Since electricity can’t be stored in large quantities for an extended period of time, hence making it relevant to predict the demand by consumers. Moreover, a peak demand forecast becomes necessary since transmission companies lose energy during peak periods. Therefore, a Kalman-filtered Monte Carlo method was implemented to forecast the peak and short-term demand.
  • Classification : 62Pxx, 60Gxx, 60Jxx, Bayesian Time Series Forecasting
  • Author(s) :
    • Frank Kofi Owusu (Kumasi Technical University)
    • Nana Kena Frempong (Kwame Nkrumah University of Science and Technology)
    • Peter Amoako Yirenkyi (Kwame Nkrumah University of Science and Technology)
    • Isaac Adjei Mensah (Kwame Nkrumah University of Science and Technology)