cover
  • Profile
  • Backtests
  • Community

Biography

I am a Computer Science student at the University of Washington. I am currently an intern at QuantConnect.

Activity on QuantConnect

We are pioneering the radical future for open-source quant finance. QuantConnect is the world's largest quant community, empowering 220,000 quants with a framework, data, and infrastructure for their investments.


Public Backtests (811)

View More
Square Fluorescent Orange Armadillo

-13.978Net Profit

2.959PSR

-1.087Sharpe Ratio

-0.195Alpha

0.64Beta

-26.058CAR

24.1Drawdown

-0.3Loss Rate

5Parameters

0Security Types

-1.663Sortino Ratio

0Tradeable Dates

96Trades

-0.278Treynor Ratio

0.05Win Rate

Energetic Fluorescent Orange Termite

0.468Net Profit

38.527PSR

0.254Sharpe Ratio

-0.05Alpha

0.357Beta

4.404CAR

8.6Drawdown

100Loss Rate

28Parameters

2Security Types

0Sortino Ratio

238Tradeable Dates

7Trades

0.2Treynor Ratio

0Win Rate

Casual Black Cat

-5.982Net Profit

4.797PSR

-3.334Sharpe Ratio

-0.453Alpha

-0.441Beta

-48.81CAR

6.6Drawdown

0Loss Rate

19Parameters

0Security Types

-4.632Sortino Ratio

24Tradeable Dates

16Trades

1.07Treynor Ratio

0Win Rate

Crying Yellow-Green Mule

-19.441Net Profit

21.131PSR

-0.794Sharpe Ratio

2.403Alpha

-6.316Beta

-84.37CAR

27.4Drawdown

0Loss Rate

20Parameters

0Security Types

-1.424Sortino Ratio

273Tradeable Dates

10001Trades

0.121Treynor Ratio

0.02Win Rate

Calculating Light Brown Hamster

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

25Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

2Tradeable Dates

10Trades

0Treynor Ratio

0Win Rate


Community

View More

Shile submitted the research SVM Wavelet Forecasting

Abstract

In this tutorial, the authors explore the application of an SVM Wavelet model for forecasting EURJPY prices. They combine a Support Vector Machine (SVM) with Wavelets to handle non-linear data and decompose the time-series into multiple components. The SVM is then applied to forecast one time-step ahead for each component, and the components are recombined to obtain the aggregate forecast. The algorithm achieved a Sharpe Ratio of 0.553, outperforming buying and holding SPY. Suggestions for improvement include testing different Wavelet types, trying out other time resolutions, and using alternative Decomposition methods. The authors encourage users to share any interesting results or modifications in the Community Forum.

3 years ago

Shile started the discussion My QuantConnect workflow (as a previous QuantConnect support staff)

Hi Everyone,

3 years ago

Shile left a comment in the discussion Strategy Library Addition: Forecasting Stock Prices using a Temporal CNN Model

Hi Carpediem911,

3 years ago

Shile left a comment in the discussion Is there a way to pass indicators from Universe Selection Model to Alpha Model?

Hi Fishstoryyy,

3 years ago

Shile left a comment in the discussion Rolling Window not working Lean Locally

Hi Ishant,

3 years ago

Square Fluorescent Orange Armadillo

-13.978Net Profit

2.959PSR

-1.087Sharpe Ratio

-0.195Alpha

0.64Beta

-26.058CAR

24.1Drawdown

-0.3Loss Rate

5Parameters

0Security Types

-1.663Sortino Ratio

0Tradeable Dates

96Trades

-0.278Treynor Ratio

0.05Win Rate

Energetic Fluorescent Orange Termite

0.468Net Profit

38.527PSR

0.254Sharpe Ratio

-0.05Alpha

0.357Beta

4.404CAR

8.6Drawdown

100Loss Rate

28Parameters

2Security Types

0Sortino Ratio

238Tradeable Dates

7Trades

0.2Treynor Ratio

0Win Rate

Casual Black Cat

-5.982Net Profit

4.797PSR

-3.334Sharpe Ratio

-0.453Alpha

-0.441Beta

-48.81CAR

6.6Drawdown

0Loss Rate

19Parameters

0Security Types

-4.632Sortino Ratio

24Tradeable Dates

16Trades

1.07Treynor Ratio

0Win Rate

Crying Yellow-Green Mule

-19.441Net Profit

21.131PSR

-0.794Sharpe Ratio

2.403Alpha

-6.316Beta

-84.37CAR

27.4Drawdown

0Loss Rate

20Parameters

0Security Types

-1.424Sortino Ratio

273Tradeable Dates

10001Trades

0.121Treynor Ratio

0.02Win Rate

Calculating Light Brown Hamster

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

25Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

2Tradeable Dates

10Trades

0Treynor Ratio

0Win Rate

Calm Fluorescent Orange Llama

7.679Net Profit

41.931PSR

0.867Sharpe Ratio

-0.371Alpha

6.394Beta

22.841CAR

46.6Drawdown

-4.22Loss Rate

42Parameters

0Security Types

0.059Sortino Ratio

92Tradeable Dates

48Trades

0.109Treynor Ratio

2.61Win Rate

Calm Red Pig

-5.627Net Profit

10.613PSR

-1.359Sharpe Ratio

-0.285Alpha

-0.169Beta

-28.892CAR

11.6Drawdown

-1.3Loss Rate

39Parameters

0Security Types

-1.161Sortino Ratio

41Tradeable Dates

33Trades

1.683Treynor Ratio

0.43Win Rate

Muscular Fluorescent Orange Falcon

-5.627Net Profit

10.613PSR

-1.359Sharpe Ratio

-0.285Alpha

-0.169Beta

-28.892CAR

11.6Drawdown

-1.3Loss Rate

39Parameters

0Security Types

-1.161Sortino Ratio

41Tradeable Dates

33Trades

1.683Treynor Ratio

0.43Win Rate

Casual Fluorescent Yellow Lion

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

42Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

7Tradeable Dates

0Trades

0Treynor Ratio

0Win Rate

Emotional Yellow-Green Chimpanzee

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

42Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

7Tradeable Dates

0Trades

0Treynor Ratio

0Win Rate

Formal Fluorescent Orange Llama

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

42Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

7Tradeable Dates

0Trades

0Treynor Ratio

0Win Rate

Determined Green Owlet

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

42Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

7Tradeable Dates

0Trades

0Treynor Ratio

0Win Rate

Determined Orange Mosquito

-80.423Net Profit

0.015PSR

-0.455Sharpe Ratio

-1.002Alpha

1.46Beta

-100CAR

81.6Drawdown

98Loss Rate

39Parameters

2Security Types

-7.117Sortino Ratio

920Tradeable Dates

156Trades

-0.685Treynor Ratio

2Win Rate

Alert Yellow-Green Bear

-13.301Net Profit

0.195PSR

-1.312Sharpe Ratio

-0.111Alpha

0.067Beta

-13.101CAR

16.3Drawdown

-0.38Loss Rate

17Parameters

0Security Types

-1.703Sortino Ratio

316Tradeable Dates

739Trades

-1.489Treynor Ratio

0.48Win Rate

Logical Fluorescent Orange Galago

25.78Net Profit

67.385PSR

2.19Sharpe Ratio

0.185Alpha

2.369Beta

72.812CAR

17.4Drawdown

0Loss Rate

13Parameters

0Security Types

2.53Sortino Ratio

130Tradeable Dates

5Trades

0.29Treynor Ratio

11.91Win Rate

Virtual Red-Orange Crocodile

-29.167Net Profit

12.016PSR

0.24Sharpe Ratio

-0.029Alpha

1.769Beta

-15.838CAR

89Drawdown

-17.89Loss Rate

11Parameters

0Security Types

-0.073Sortino Ratio

620Tradeable Dates

10Trades

0.088Treynor Ratio

17.53Win Rate

Focused Red Rhinoceros

396.383Net Profit

90.566PSR

6.932Sharpe Ratio

5.998Alpha

2.288Beta

395.795CAR

34.1Drawdown

-2.43Loss Rate

38Parameters

0Security Types

5.651Sortino Ratio

253Tradeable Dates

209Trades

2.776Treynor Ratio

3.61Win Rate

Well Dressed Fluorescent Yellow Beaver

12.008Net Profit

37.144PSR

0.826Sharpe Ratio

0.024Alpha

0.083Beta

3.718CAR

5.4Drawdown

-0.66Loss Rate

34Parameters

0Security Types

0.723Sortino Ratio

780Tradeable Dates

200Trades

0.465Treynor Ratio

1.02Win Rate

Hyper-Active Orange Dinosaur

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

30Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

277Tradeable Dates

0Trades

0Treynor Ratio

0Win Rate

Crawling Fluorescent Yellow Buffalo

0Net Profit

0PSR

0Sharpe Ratio

0Alpha

0Beta

0CAR

0Drawdown

0Loss Rate

20Parameters

0Security Types

7.9228162514264E+28Sortino Ratio

10Tradeable Dates

0Trades

0Treynor Ratio

0Win Rate

Square Red Koala

47.893Net Profit

59.821PSR

1.226Sharpe Ratio

0.094Alpha

0.268Beta

13.427CAR

9.6Drawdown

-2.4Loss Rate

20Parameters

0Security Types

1.109Sortino Ratio

780Tradeable Dates

34Trades

0.531Treynor Ratio

5.89Win Rate

Geeky Sky Blue Donkey

55.499Net Profit

10.618PSR

0.51Sharpe Ratio

0.101Alpha

-0.05Beta

9.223CAR

31.4Drawdown

-0.41Loss Rate

16Parameters

0Security Types

0.517Sortino Ratio

1259Tradeable Dates

1712Trades

-1.864Treynor Ratio

0.4Win Rate

Hyper-Active Fluorescent Orange Hyena

91.721Net Profit

7.509PSR

0.516Sharpe Ratio

0.118Alpha

-0.05Beta

10.528CAR

39.2Drawdown

-0.41Loss Rate

16Parameters

1Security Types

0.513Sortino Ratio

1637Tradeable Dates

2156Trades

-2.207Treynor Ratio

0.54Win Rate

Emotional Orange Wolf

68.553Net Profit

4.56PSR

0.419Sharpe Ratio

0.101Alpha

-0.023Beta

8.36CAR

42.2Drawdown

-0.49Loss Rate

16Parameters

1Security Types

0.334Sortino Ratio

1637Tradeable Dates

1966Trades

-4.22Treynor Ratio

0.52Win Rate

Hipster Blue Lemur

69.069Net Profit

4.551PSR

0.418Sharpe Ratio

0.103Alpha

-0.025Beta

8.411CAR

42.6Drawdown

-0.5Loss Rate

15Parameters

1Security Types

0.33Sortino Ratio

1637Tradeable Dates

1932Trades

-3.929Treynor Ratio

0.53Win Rate

Retrospective Orange Hyena

179.391Net Profit

76.855PSR

2.156Sharpe Ratio

1.439Alpha

-0.926Beta

91.552CAR

21.1Drawdown

27Loss Rate

32Parameters

2Security Types

0.8525Sortino Ratio

398Tradeable Dates

56Trades

-1.279Treynor Ratio

73Win Rate

Calm Light Brown Bison

-3.333Net Profit

7.576PSR

-3.387Sharpe Ratio

-0.157Alpha

0.388Beta

-32.066CAR

5.8Drawdown

0Loss Rate

34Parameters

0Security Types

-4.897Sortino Ratio

20Tradeable Dates

3Trades

-0.853Treynor Ratio

0Win Rate

Hyper-Active Yellow Koala

-34.182Net Profit

2.565PSR

-0.618Sharpe Ratio

0.006Alpha

0.916Beta

-26.913CAR

49.4Drawdown

-0.24Loss Rate

27Parameters

0Security Types

-1.144Sortino Ratio

336Tradeable Dates

21Trades

-0.24Treynor Ratio

0Win Rate

Focused Sky Blue Cobra

17.436Net Profit

74.799PSR

2.237Sharpe Ratio

0.074Alpha

0.72Beta

33.63CAR

7.9Drawdown

40Loss Rate

28Parameters

2Security Types

-0.6191Sortino Ratio

645Tradeable Dates

11Trades

0.509Treynor Ratio

60Win Rate

Calculating Yellow-Green Armadillo

44.871Net Profit

54.325PSR

1.248Sharpe Ratio

0.057Alpha

0.867Beta

26.792CAR

27.7Drawdown

0Loss Rate

9Parameters

1Security Types

0Sortino Ratio

393Tradeable Dates

13Trades

0.354Treynor Ratio

100Win Rate

Shile submitted the research SVM Wavelet Forecasting

Abstract

In this tutorial, the authors explore the application of an SVM Wavelet model for forecasting EURJPY prices. They combine a Support Vector Machine (SVM) with Wavelets to handle non-linear data and decompose the time-series into multiple components. The SVM is then applied to forecast one time-step ahead for each component, and the components are recombined to obtain the aggregate forecast. The algorithm achieved a Sharpe Ratio of 0.553, outperforming buying and holding SPY. Suggestions for improvement include testing different Wavelet types, trying out other time resolutions, and using alternative Decomposition methods. The authors encourage users to share any interesting results or modifications in the Community Forum.

3 years ago

Shile started the discussion My QuantConnect workflow (as a previous QuantConnect support staff)

Hi Everyone,

3 years ago

Shile left a comment in the discussion Strategy Library Addition: Forecasting Stock Prices using a Temporal CNN Model

Hi Carpediem911,

3 years ago

Shile left a comment in the discussion Is there a way to pass indicators from Universe Selection Model to Alpha Model?

Hi Fishstoryyy,

3 years ago

Shile left a comment in the discussion Rolling Window not working Lean Locally

Hi Ishant,

3 years ago

Shile left a comment in the discussion Need help with quarterly or biweekly rebalancing in Algorithm Framework

Hi Rakesh,

3 years ago

Shile left a comment in the discussion Rolling Window of Consolidated bars

Hi Luke,

3 years ago

Shile left a comment in the discussion Implementing Black-Litterman Port Construction with EMA, SMA and MACD Alphas

Hi Miguel,

3 years ago

Shile submitted the research Forecasting Stock Prices Using A Temporal CNN Model

Abstract

This tutorial focuses on using a Temporal Convolutional Neural Network (CNN) model to forecast stock prices. The goal is to develop a model that can predict the movement of future prices based on historical data. The input to the model is OHLC+Volume data for the past 15 time steps, and the output is the direction of the movement of the average close for the next 5 time steps. The model is built using Keras, a Python Deep Learning API. The code required to build the Neural Network Architecture and prepare the data is explained in detail. The tutorial also mentions that the algorithm is non-deterministic, so different results may be seen in repeated backtests.

4 years ago

Shile submitted the research Leveraged ETFs With Systematic Risk Management

Abstract

This tutorial explores the use of leverage and systematic risk management to outperform the S&P500 Index. The authors propose using leveraged S&P500 indices with systematic risk management, utilizing moving averages to assess market volatility and manage risk. The method involves holding or liquidating positions based on the 200-day Simple Moving Average (SMA) of the SSO ETF, a 2x leveraged S&P500 index ETF. If the current price of SSO is above the 200-day SMA, the position is held, and if it dips below, the position is liquidated and rotated into short-term treasuries using the SHY ETF. The results show a higher Sharpe Ratio for this strategy compared to buying and holding the SPY ETF.

4 years ago

Shile submitted the research Optimal Pairs Trading

Abstract

Abstract: This tutorial explores the application of the Ornstein-Uhlenbeck model to pairs trading and the derivation of optimal entry and liquidation levels. Pairs trading involves holding one stock while shorting another stock in order to profit from the convergence of their spread. The tutorial discusses the use of a Kalman Filter for execution and proposes modeling the portfolio values as an Ornstein-Uhlenbeck process. The tutorial outlines the method for computing the Ornstein-Uhlenbeck coefficients and selecting the optimal beta value. It also explains how to derive the optimal entry and liquidation levels based on the coefficients. The implementation details can be found in the provided code.

4 years ago

Shile submitted the research G Score Investing

Abstract

This tutorial discusses G-Score Investing, a method of trading that involves analyzing a company's fundamentals to make investment decisions. The tutorial explains the concept of Factor Investing and how G-Score investing evaluates companies based on seven factors, with a focus on Book-to-Market ratio. The method involves sorting companies by their Book-to-Market ratio and selecting the bottom quartile. The tutorial also mentions that fundamental data used in the algorithms is sourced from MorningStar. The results of the algorithm are compared to the performance of the QQQ ETF as a benchmark. The tutorial concludes by referencing a paper by Mohanram on separating winners from losers among low Book-to-Market stocks using financial statement analysis.

4 years ago

Shile started the discussion My Pairs Trading Algorithm has a near perfect linear downward trend

I found this pretty interesting. If someone finds a way to take advantage of this, please let me...

4 years ago

Shile started the discussion Strategy Library Addition: Leveraged ETFs with Systematic Risk Management

Hi Everyone,

4 years ago

Shile started the discussion Strategy Library Addition: Forecasting Stock Prices using a Temporal CNN Model

Hi Everyone,

4 years ago