Lulu Wang

Realized Volatility Prediction and Graph Data Mining

Graph Neural Networks(GNNs) are very powerful, but what if the graph is nonobservable?
We proposed a new model integrating graph discovery data mining techniques and GNNs to overcome limitations in graph structures for machine learning modeling tasks, and applied it to realized Volatility Prediction.

• Learn more
• Dashboard

Recommender System for Advertising

If a supplement brand and gym wear brand hired the same influencer, does it help if they hire another together?
We developed a Graph Neural Networks (GNNs) based recommender system connecting businesses with social media influencers by learning various aspects of their social media behaviors such as text caption, image, and the social relationship among influencers and brands.

• Learn more

Agent-based Model Parameters Estimation with Wasserstein Distance

The goal of this paper is to propose an estimation method that is accurate when the interest is to estimate the parameters of high-dimensional models, such as DSGE and ABM. In particular, we propose a Simulated Minimum Distance (SMD) estimator based on the Wasserstein distance between the model-simulated distribution and the empirical ...

• Learn more
• Slides

Surrogate Models for Calibration

In this project, we experimented few popular machine learning regression and classification models as surrogate models to calibrate the Brock and Hommes asset pricing model using S&P 500 price time series data.

• Poster

Factor models

Factor modeling comprises three frameworks:

• Factors are observable but their exposures are latent. Examples include Fama-French factor models.
• Factor exposures are observable, but the factors are latent. Illustrated by MSCI Barra model.
• Factors and their exposures are latent.

We formulated a detailed summary paper for the Barra-type factor structure.

• slides
• paper

Forecasting Emerging Markets Currencies with Financial Data

Belonged to the third framework of factor modeling: factors and their exposures are latent, this paper uses financial variables to simulate and forecast short-run emerging markets (EM) exchange rates and provide empirical investment strategies with sound back-test returns. We ...

• Learn moree
• Paper

Endogenous Demand Driven Supply Chain Cycles with Exogenous Shocks

In this article, we model the inventories and productions of a demand driven supply chain consisting of upstream, midstream, downstream and retail market. The dynamic system is a network of two-dimensional oscillators, considering positive feedback from lower stream demand and negative ...

• Learn more
• Paper

Comparing Analysis of the United States and Germany Pension System Reform

It has been more than 20 years since the World Bank’s influential report suggested pension reform and a multi-pillar system for the provision of old-age income security. However, not all country's attempts toward such goals are met by success. The U.S.’s pension system reform was wildly regarded as ...

• Learn more
• Paper

China’s Experience in 2015 Stock Market Crash

In this article, I introduce China's state-owned financial institutions and their actions in intervening in the stock market during the 2015 Chinese stock crash and summarize some policy impact studies of this matter. Overall, the intervention had a positive effect on stabilizing the stock market. Rescue funds also outperformed the benchmark during their stock holding period.

• Paper