Pace University CS677-Introduction to Machine Learning Spring 2022 Computer Science Department Prof. Sarbanes Page 1 of 2 Computer Science Department CS677 Introduction to Machine Learning (CRN: 27964) Spring 2022 Project #1 / Due 01-Mar-2022(*) The goal of this assignment is to understand the logic, methods, and types available to implement machine learning classifiers. Classification is the process of predicting the class of given data points. Classes are sometimes called as targets/ labels or categories. A classifier is the algorithm itself the rules used by machines to classify data. A classification model, on the other hand, is the end result of your classifiers machine learning. The model is trained using the classifier, so that the model, ultimately, classifies your data. In this project, well deal with specific type of data input, financial data. Specifically, well try to build a trading (security) strategy for future stock purchases based on historical values! You should get the historical data (this is real/production data) from a credible source, see below. Prior commencing your efforts on coding, you must install/utilize the following libraries: ? Yahoo Finance, or ? GoogleFinance API https://pypi.org/project/yfinance/ https://support.google.com/docs/answer/3093281 There are many more finance (stock prices) alternative data sources such as Quandl, Interactive Brokers, Alpha Vantage, TradingView, IEX, etc I would highly recommend to use the Yahoo Finance (yfinance) Python library. Get data for any security of your choice (AAPL, IBM, GOOG, or the S&P500 index SPY) from beginning of 2015 until the end of last year. For example, to download data for SPY execute: spy_data = yf.download(\"SPY\", start=\"2015-01-01\", end=\"2021-12-31\") Perform two (2) distinct simple trading strategies in Python which will be based on Machine Learning Classification modeling approaches. The strategies will be such that will call for a buy or sell signal of a certain security. The signal will equate to the target/label variable used by the ML classifiers. Namely, the target variable would be either a buying or selling the stock based on feature variables. The target variable content should be constructed as follows: Store the value of +1 for the buy signal and -1 for the sell signal, respectively. Strategy-1: If the next trading day\'s close price is greater than today\'s close price then, the signal is buy, otherwise sell. Strategy-2: Utilize the 50-day moving average vs the 200-day moving average. A golden cross (or golden crossover) is a chart pattern that involves a short-term moving average crossing above a long-term moving average. Typically, the 50-day MA is used as the short-term average, and the 200-day MA is used as the long-term average. This is an indicator of bullish (buying) signal. (See Figure-1 next page) A death cross is basically the opposite of a golden cross. Its a chart pattern where a short-term MA crosses below a long-term MA. For example, the 50-day MA crosses below the 200-day MA. As such, a death cross is typically considered to be a bearish (selling) signal. Project #1 Sarbanes -- Page 2 of 2 Figure-1: Golden Cross (Bullish Signal) NumPy library provides a plethora of building methods to handle current (todays) price, yesterdays price, Moving Averages (Mas), Roll, etc. For instance, to calculate the values of the target variable for Strategy-1, use this NumPy call: y = np.where(df[Stock_Price].shift(-1) > df[Stock_Price], 1, -1) For Moving Averages (Strategy-2_, you should come up with a similar formula. The following steps, generating the ML classifiers, should be performed: ? Pre-process and clean the data. ? Define the Feature Variable X, and the Label/Target variable y. ? Spilt the data into training and test datasets (use the 8020 percent ratio.) ? Choose a Classifier and fit it on the training dataset (take its default parameters) ? Evaluate the Classifier on the test dataset. Utilize the following ML Classifier Models: 1_ K-Nearest Neighbors (KNN) from sklearn.neighbors import KNeighborsClassifier 2_ Random Forest Classifier (RF) from sklearn.ensemble import RandomForestClassifier 3_ Gradient Boosting Classifier (GB) from sklearn.ensemble import GradientBoostingClassifier 4_ Support Vector Machines (SVMs) from sklearn.svm import SVC 5_ XGBoost Classifier (Note: You don\'t import it from scikit-learn module!) (You need to install it, it not available: pip install xgboost) from xgboost import XGBClassifier Write Python scripts in order to complete the above tasks along with their output. All work should be done and submitted in a single Jupyter (or Colab) Notebook. Extra Credit: Try to tune some classifiers (you decide which ones, ever 1 will do), by changing their default parameters.