Skip to content
View garroshub's full-sized avatar

Block or report garroshub

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
garroshub/README.md

Hi there, I'm Garros Gong

Linkedin Badge Website Badge GitHub Views

GIF

Glad to see you here! Β 

I love to talk data science, technology, investment and sustainability topics.

My digital portfolio

Streamlit App

Talking about Personal Stuffs:

  • πŸ‘¨πŸ»β€πŸ’» I’m currently working in the financnial industry;
  • πŸš€ I’m currently doing my PhD in Management Sciences: Applied Operational Research (part-time basis)

πŸ“ˆ My GitHub Stats:

πŸ“š Academic Publications

  • Authors: Garros Gong, Stanko Dimitrov, Michael R. Bartolacci
  • Journal: Discover Sustainability
  • Year: 2024
  • Abstract: This study proposes the integration of specific social media analytics (SMA) metrics into existing U.S. wildfire management systems to enhance their ability to accurately predict, monitor, and respond to wildfires in a timely manner. In addition, the examination of SMA's influence on shaping wildfire-related policies is addressed in our analysis with respect to the mitigation of the extent and effects of such disasters. Furthermore, the potential of Web 3.0 technologies in achieving these objectives is analyzed as part of this work. The results highlight that advaa analytics (SMA) metrics to wildfire management and along with Web 3.0 integration.
  • Keywords: wildfire management, social media analytics metrics, Web 3.0, policymaking

πŸ“– Working Papers

  • Authors: Garros Gong, Stanko Dimitrov, Michael R. Bartolacci
  • Journal: SSRN
  • Year: 2024
  • Abstract: This study investigates the impact of federal budget allocations for wildfire management on the accuracy of wildfire preparedness decisions across different budget cycles, providing evidence for fiscal policy non-neutrality in this context. By analyzing a unique cross-sectional panel dataset that integrates government wildfire reports, fiscal data, and social media activity, we develop two testable hypotheses, focusing on whether social media attention and budgetary shocks influence wildfire preparedness decisions. The results indicate that social media attention (H1) positively impacts the accuracy of wildfire preparedness, albeit this effect diminishes over time. Meanwhile, positive budget shocks (H2) enhance accuracy in the short term within a single budget cycle. However, these effects are not sustained across multiple cycles, where rapid budget increases may lead to long-term inefficiencies and increased volatility in decision accuracy. These results provide empirical evidence supporting Post-Keynesian views on fiscal policy importance and longterm uncertainty in the context of wildfire management, highlighting the need for carefully designed budget plans and caution regarding aggressive green policy changes. The study also offers insights into social media use during wildfires, calling for customized strategies targeted at less populated areas.
  • Keywords: Post-Keynesian, Sustainable Budget Plan, Green Fiscal Policy, Social Media Strategy, Wildfire Management
  • Authors: Garros Gong, Stanko Dimitrov, Michael R. Bartolacci
  • Journal: SSRN
  • Year: 2024
  • Abstract: This study examines the micro-level drivers of wildfire economic costs by analyzing both endogenous factors, such as preparedness levels, total physical resources assigned, and other features reported in government wildfire reports, as well as exogenous factors, particularly Twitter mentions. Using data from California wildfires from 2006 to 2021 and correlated tweets, we designed three testable hypotheses to identify significant endogenous and exogenous factors and determine the magnitude of their influence on the economic outcomes of wildfires. The first part of hypothesis testing showed preparedness levels and total physical resources assigned significantly connected to wildfire economic costs, also confirming that informed planning on preparedness and resource allocation better reduces total economic costs than increasing emergency efforts during wildfires. The second part of hypothesis testing recognized that social media activity did not directly affect total economic costs but influenced economic outcomes through interactions with preparedness levels and total physical resources assigned. Notably, a high volume of tweets about wildfires indirectly drove higher total economic costs. These results suggest that government agencies should avoid letting high social media attention dictate changes in planning and resource allocation, as this could potentially undermine efforts to mitigate economic loss from wildfires.
  • Keywords: Wildfire Economics, Social Media Analytics, California Wildfires, Active Wildfire Management

Pinned Loading

  1. Modern-Portfolio-Optimizer Modern-Portfolio-Optimizer Public

    This Streamlit app calculates an optimal investment portfolio based on a user-defined minimum required return and selection of funds.

    Python

  2. ESG_Analysis_NLP ESG_Analysis_NLP Public

    Are you interested in ESG investment? Here is a sample project that deploys cutting-edge NLP techniques to analyze a company's ESG performance.

    Python

  3. Equity_Research_Auto Equity_Research_Auto Public

    This app provides a comprehensive equity research dashboard using data from Financial Modeling Prep API.

    Python

  4. Option-Trading-Analytical-Platform-App Option-Trading-Analytical-Platform-App Public

    Purpose: Compare different option chains and calculate value of calls/puts with Black Scholes

    Python 1

  5. Stock_Price_Prediction_LSTM-Deep-Learning Stock_Price_Prediction_LSTM-Deep-Learning Public

    Predicting closing stock prices using Deep Learning models such as Long Short Term Memory (LSTM), a Basic Artificial Neural Network(CNN), Recurrent Neural Networks (RNN), Multilayer Perceptron (MLP…

    Python

  6. Credit_Card_Fraud_Detection Credit_Card_Fraud_Detection Public

    A machine learning model using classification algorithms and techniques to accurately detect if a credit card transaction is fraudulent or not.

    Python