SparshKumar_Impl_1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6cc0a6d0",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow in /Users/spars/anaconda3/lib/python3.11/site-packages (2.14.0)\n",
      "Requirement already satisfied: tensorflow-macos==2.14.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow) (2.14.0)\n",
      "Requirement already satisfied: absl-py>=1.0.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (2.0.0)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (1.6.3)\n",
      "Requirement already satisfied: flatbuffers>=23.5.26 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (23.5.26)\n",
      "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (0.5.4)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (0.2.0)\n",
      "Requirement already satisfied: h5py>=2.9.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (3.7.0)\n",
      "Requirement already satisfied: libclang>=13.0.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (16.0.6)\n",
      "Requirement already satisfied: ml-dtypes==0.2.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (0.2.0)\n",
      "Requirement already satisfied: numpy>=1.23.5 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (1.24.3)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (3.3.0)\n",
      "Requirement already satisfied: packaging in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (23.0)\n",
      "Requirement already satisfied: protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (4.24.4)\n",
      "Requirement already satisfied: setuptools in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (68.0.0)\n",
      "Requirement already satisfied: six>=1.12.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (1.16.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (2.3.0)\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (4.7.1)\n",
      "Requirement already satisfied: wrapt<1.15,>=1.11.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (1.14.1)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (0.34.0)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (1.59.0)\n",
      "Requirement already satisfied: tensorboard<2.15,>=2.14 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (2.14.1)\n",
      "Requirement already satisfied: tensorflow-estimator<2.15,>=2.14.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (2.14.0)\n",
      "Requirement already satisfied: keras<2.15,>=2.14.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorflow-macos==2.14.0->tensorflow) (2.14.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from astunparse>=1.6.0->tensorflow-macos==2.14.0->tensorflow) (0.38.4)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (2.23.3)\n",
      "Requirement already satisfied: google-auth-oauthlib<1.1,>=0.5 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (1.0.0)\n",
      "Requirement already satisfied: markdown>=2.6.8 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (3.4.1)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (2.31.0)\n",
      "Requirement already satisfied: tensorboard-data-server<0.8.0,>=0.7.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (0.7.1)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (2.2.3)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (5.3.1)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (0.2.8)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in /Users/spars/anaconda3/lib/python3.11/site-packages (from google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from google-auth-oauthlib<1.1,>=0.5->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (1.3.1)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /Users/spars/anaconda3/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (2.0.4)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /Users/spars/anaconda3/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (1.26.16)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /Users/spars/anaconda3/lib/python3.11/site-packages (from requests<3,>=2.21.0->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (2023.7.22)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in /Users/spars/anaconda3/lib/python3.11/site-packages (from werkzeug>=1.0.1->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (2.1.1)\n",
      "Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /Users/spars/anaconda3/lib/python3.11/site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (0.4.8)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in /Users/spars/anaconda3/lib/python3.11/site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<1.1,>=0.5->tensorboard<2.15,>=2.14->tensorflow-macos==2.14.0->tensorflow) (3.2.2)\n"
     ]
    }
   ],
   "source": [
    "!pip install tensorflow\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "05fe692d",
   "metadata": {},
   "source": [
    "## Features Chosen\n",
    "\n",
    "Candlestick patterns are a common tool used in technical analysis of stock. Candlesticks encapsulates 4 data points of a stock over a given point in time - Open, Close, High and Low. By tracking candlesticks over a period of time, traders try to find candlestick patterns which signal future price fluctuations. Often, Volume data is incorporated into this analysis (as increased trading volume is a signal of violatility). \n",
    "\n",
    "Therefore, I chose to incorporate NVDA's daily Open, Close, High, Low and Volume data as a time series to replicate the technical analysis performed using candlestick patterns.\n",
    "\n",
    "However, I also wanted to include wider market and industry (beta) movements as a feature. This is because NVDA's stock price may lag external shocks by a day or two. Recent stock market performance informed my decision to choose the daily Nasdaq 100 index (NDX) close price to measure the influence of the wider market. I did not choose the S&P 500 due to the large uncoupling that has taken place between the technology sector and the overall market over the past two years. \n",
    "\n",
    "Final features - NVDA Open, Close, High, Low, Volume and NDX Close (in a timeseries). I decided to utilise a sliding window method which looks at the last 60 days of trading data and predicts the next 10 days.\n",
    "\n",
    "## Model Chosen\n",
    "\n",
    "As my dataset was in a time series, I chose an LSTM model due to it's strength in modeling long-term dependency and its ability to selectively forget irrelevant data as I imagined my dataset would have a lot of noise. \n",
    "\n",
    "However, as I discuss later down, I found an neural network of only LSTM layers was prone to overfitting to this particular dataset and eventually decided to implement a single LSTM layer followed by several feed-forward Dense layers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "8a0a5b80",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow import keras\n",
    "import numpy as np\n",
    "import pandas\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from datetime import datetime\n",
    "from sklearn.metrics import mean_squared_error\n",
    "from math import sqrt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4237944e",
   "metadata": {},
   "outputs": [],
   "source": [
    "NUMBER_OF_DAYS = 11  # Number of days price is predicted for (based on project parameters)\n",
    "WINDOW = 60  \n",
    "slice_date = 1500  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3a01fdca",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                  Open        High         Low       Close     Volume  \\\n",
      "Date                                                                    \n",
      "2005-01-10    1.841667    1.858333    1.829167    1.840000   57955200   \n",
      "2005-01-11    1.822500    1.832500    1.755833    1.783333  109338000   \n",
      "2005-01-12    1.799167    1.805833    1.725000    1.768333  105780000   \n",
      "2005-01-13    1.772500    1.833333    1.733333    1.786667   89876400   \n",
      "2005-01-14    1.820000    1.824167    1.775000    1.792500   57391200   \n",
      "...                ...         ...         ...         ...        ...   \n",
      "2023-10-17  440.000000  447.540009  424.799988  439.380005   81233300   \n",
      "2023-10-18  425.910004  432.190002  418.250000  421.959991   62729400   \n",
      "2023-10-19  428.109985  432.970001  418.820007  421.010010   50123300   \n",
      "2023-10-20  418.899994  424.700012  410.779999  413.869995   47638100   \n",
      "2023-10-23  412.290009  432.480011  409.450012  429.750000   47721800   \n",
      "\n",
      "               NDX_Close  \n",
      "Date                      \n",
      "2005-01-10   1564.920044  \n",
      "2005-01-11   1553.400024  \n",
      "2005-01-12   1565.780029  \n",
      "2005-01-13   1545.180054  \n",
      "2005-01-14   1561.109985  \n",
      "...                  ...  \n",
      "2023-10-17  15122.009766  \n",
      "2023-10-18  14909.259766  \n",
      "2023-10-19  14783.129883  \n",
      "2023-10-20  14560.879883  \n",
      "2023-10-23  14604.849609  \n",
      "\n",
      "[4729 rows x 6 columns]\n"
     ]
    }
   ],
   "source": [
    "# Import NVDA Stock Data\n",
    "ndva = pandas.read_csv(\"NVDA.csv\")\n",
    "ndva.set_index('Date', inplace=True)\n",
    "ndva_first_index = ndva.index[0]\n",
    "\n",
    "# Import NDX Index Data and match index to NVDA data\n",
    "ndx = pandas.read_csv(\"^NDX.csv\")\n",
    "ndx.set_index('Date', inplace=True)\n",
    "ndx = ndx.loc[ndva_first_index:]\n",
    "ndx_filtered = ndx['Close']\n",
    "\n",
    "# Combine NVDA and NDX data\n",
    "data = ndva.copy()\n",
    "data['NDX_Close'] = ndx_filtered\n",
    "\n",
    "# Choose features \n",
    "FEATURES = ['Open', 'High', 'Low', 'Close', 'Volume', 'NDX_Close']\n",
    "data_filtered = data[FEATURES]\n",
    "data_filtered = data_filtered.iloc[slice_date:]\n",
    "print(data_filtered)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f08d59af",
   "metadata": {},
   "outputs": [],
   "source": [
    "np_data = np.array(data_filtered)\n",
    "# Scale dataset\n",
    "scaler = MinMaxScaler()\n",
    "np_data_scaled = scaler.fit_transform(np_data)\n",
    "\n",
    "# Seperate scaler to unscale predictions later\n",
    "prediction_scaler = MinMaxScaler()\n",
    "close_data = np.reshape(np.array(data_filtered['Close']), (-1, 1))\n",
    "close_scaled = prediction_scaler.fit_transform(close_data)\n",
    "\n",
    "index_Close = data_filtered.columns.get_loc(\"Close\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a8fef005",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split training and test data\n",
    "training_test_split = 0.7\n",
    "split_data = int(np.ceil(np_data_scaled.shape[0] * training_test_split))\n",
    "test_start_date = data_filtered.index[split_data]\n",
    "training_data = np_data_scaled[0:split_data, :]\n",
    "test_data = np_data_scaled[split_data - WINDOW:, :]\n",
    "\n",
    "\n",
    "def add_dimension(window, data):\n",
    "    x = []\n",
    "    y = []\n",
    "\n",
    "    for i in range(window, data.shape[0] - NUMBER_OF_DAYS):\n",
    "        x.append(data[i - window:i, :])\n",
    "        y.append(data[i:i + NUMBER_OF_DAYS, index_Close])\n",
    "\n",
    "    x = np.array(x)\n",
    "    y = np.array(y)\n",
    "    return x, y\n",
    "\n",
    "\n",
    "x_train, y_train = add_dimension(WINDOW, training_data)\n",
    "x_test, y_test = add_dimension(WINDOW, test_data)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "47517b22",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " lstm (LSTM)                 (None, 360)               528480    \n",
      "                                                                 \n",
      " dense (Dense)               (None, 32)                11552     \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 32)                0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 16)                528       \n",
      "                                                                 \n",
      " dropout_1 (Dropout)         (None, 16)                0         \n",
      "                                                                 \n",
      " dense_2 (Dense)             (None, 11)                187       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 540747 (2.06 MB)\n",
      "Trainable params: 540747 (2.06 MB)\n",
      "Non-trainable params: 0 (0.00 Byte)\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Hyperparamters\n",
    "intial_nodes = x_train.shape[1] * x_train.shape[2]\n",
    "dropout_rate = 0.2\n",
    "optimiser = 'adam'\n",
    "validation_split = 0.1\n",
    "epochs = 60 \n",
    "batch_size = 32 #Tested 1, 16, 32\n",
    "\n",
    "# Build model\n",
    "model = keras.models.Sequential()\n",
    "\n",
    "model.add(keras.layers.LSTM(units=intial_nodes,\n",
    "                            return_sequences=False,\n",
    "                            input_shape=(x_train.shape[1], x_train.shape[2])))\n",
    "model.add(keras.layers.Dense(32))\n",
    "model.add(keras.layers.Dropout(dropout_rate))\n",
    "model.add(keras.layers.Dense(16))\n",
    "model.add(keras.layers.Dropout(dropout_rate))\n",
    "model.add(keras.layers.Dense(NUMBER_OF_DAYS))\n",
    "model.compile(optimizer=optimiser, loss='mse')\n",
    "model.summary()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "bcd53fc5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/60\n",
      "92/92 [==============================] - 6s 65ms/step - loss: 1.6929e-06 - val_loss: 5.0811e-05\n",
      "Epoch 2/60\n",
      "92/92 [==============================] - 6s 64ms/step - loss: 1.6048e-06 - val_loss: 5.0453e-05\n",
      "Epoch 3/60\n",
      "92/92 [==============================] - 6s 65ms/step - loss: 1.5351e-06 - val_loss: 4.1186e-05\n",
      "Epoch 4/60\n",
      "92/92 [==============================] - 6s 65ms/step - loss: 1.5884e-06 - val_loss: 2.8088e-05\n",
      "Epoch 5/60\n",
      "92/92 [==============================] - 6s 65ms/step - loss: 1.6441e-06 - val_loss: 4.2211e-05\n",
      "Epoch 6/60\n",
      "92/92 [==============================] - 6s 65ms/step - loss: 1.5814e-06 - val_loss: 2.6016e-05\n",
      "Epoch 7/60\n",
      "92/92 [==============================] - 6s 66ms/step - loss: 1.7999e-06 - val_loss: 9.5650e-05\n",
      "Epoch 8/60\n",
      "92/92 [==============================] - 6s 65ms/step - loss: 1.6778e-06 - val_loss: 5.8973e-05\n",
      "Epoch 9/60\n",
      "92/92 [==============================] - 6s 66ms/step - loss: 1.5426e-06 - val_loss: 3.0963e-05\n",
      "Epoch 10/60\n",
      "92/92 [==============================] - 6s 67ms/step - loss: 1.6888e-06 - val_loss: 3.2208e-05\n",
      "Epoch 11/60\n",
      "92/92 [==============================] - 6s 67ms/step - loss: 1.7060e-06 - val_loss: 5.1932e-05\n",
      "Epoch 12/60\n",
      "92/92 [==============================] - 6s 67ms/step - loss: 1.7328e-06 - val_loss: 3.2794e-05\n",
      "Epoch 13/60\n",
      "92/92 [==============================] - 6s 67ms/step - loss: 1.5810e-06 - val_loss: 1.2779e-04\n",
      "Epoch 14/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5523e-06 - val_loss: 3.4980e-05\n",
      "Epoch 15/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6385e-06 - val_loss: 5.0438e-05\n",
      "Epoch 16/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.7261e-06 - val_loss: 6.6403e-05\n",
      "Epoch 17/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5528e-06 - val_loss: 3.1740e-05\n",
      "Epoch 18/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.4675e-06 - val_loss: 2.5278e-05\n",
      "Epoch 19/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6214e-06 - val_loss: 5.5216e-05\n",
      "Epoch 20/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6921e-06 - val_loss: 4.7638e-05\n",
      "Epoch 21/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6429e-06 - val_loss: 1.7104e-04\n",
      "Epoch 22/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.7239e-06 - val_loss: 2.9511e-05\n",
      "Epoch 23/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5286e-06 - val_loss: 5.3754e-05\n",
      "Epoch 24/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.4814e-06 - val_loss: 4.5593e-05\n",
      "Epoch 25/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5001e-06 - val_loss: 5.3563e-05\n",
      "Epoch 26/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.8221e-06 - val_loss: 3.8595e-05\n",
      "Epoch 27/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5267e-06 - val_loss: 3.3374e-05\n",
      "Epoch 28/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5136e-06 - val_loss: 2.6634e-05\n",
      "Epoch 29/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6569e-06 - val_loss: 2.4573e-05\n",
      "Epoch 30/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.4779e-06 - val_loss: 8.7750e-05\n",
      "Epoch 31/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.3983e-06 - val_loss: 2.6187e-05\n",
      "Epoch 32/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5637e-06 - val_loss: 3.8983e-05\n",
      "Epoch 33/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6591e-06 - val_loss: 7.4155e-05\n",
      "Epoch 34/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.4159e-06 - val_loss: 2.5515e-05\n",
      "Epoch 35/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.4956e-06 - val_loss: 3.6613e-05\n",
      "Epoch 36/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 2.1632e-06 - val_loss: 1.4006e-04\n",
      "Epoch 37/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5139e-06 - val_loss: 6.7052e-05\n",
      "Epoch 38/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5719e-06 - val_loss: 3.0642e-05\n",
      "Epoch 39/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.7389e-06 - val_loss: 2.5426e-05\n",
      "Epoch 40/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.4278e-06 - val_loss: 3.2587e-05\n",
      "Epoch 41/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5750e-06 - val_loss: 2.8933e-05\n",
      "Epoch 42/60\n",
      "92/92 [==============================] - 6s 69ms/step - loss: 1.5303e-06 - val_loss: 3.3368e-05\n",
      "Epoch 43/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6364e-06 - val_loss: 9.6484e-05\n",
      "Epoch 44/60\n",
      "92/92 [==============================] - 6s 67ms/step - loss: 1.4845e-06 - val_loss: 7.2154e-05\n",
      "Epoch 45/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.8701e-06 - val_loss: 4.8355e-05\n",
      "Epoch 46/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5678e-06 - val_loss: 6.9728e-05\n",
      "Epoch 47/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6197e-06 - val_loss: 3.0702e-05\n",
      "Epoch 48/60\n",
      "92/92 [==============================] - 6s 69ms/step - loss: 1.4509e-06 - val_loss: 4.7207e-05\n",
      "Epoch 49/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.7063e-06 - val_loss: 3.4056e-05\n",
      "Epoch 50/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6476e-06 - val_loss: 2.5119e-05\n",
      "Epoch 51/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5656e-06 - val_loss: 3.4297e-05\n",
      "Epoch 52/60\n",
      "92/92 [==============================] - 6s 69ms/step - loss: 1.7819e-06 - val_loss: 2.7142e-05\n",
      "Epoch 53/60\n",
      "92/92 [==============================] - 6s 70ms/step - loss: 1.9660e-06 - val_loss: 3.1974e-05\n",
      "Epoch 54/60\n",
      "92/92 [==============================] - 6s 70ms/step - loss: 1.5529e-06 - val_loss: 2.5466e-05\n",
      "Epoch 55/60\n",
      "92/92 [==============================] - 6s 70ms/step - loss: 1.7813e-06 - val_loss: 2.5674e-05\n",
      "Epoch 56/60\n",
      "92/92 [==============================] - 6s 70ms/step - loss: 1.7961e-06 - val_loss: 2.8274e-05\n",
      "Epoch 57/60\n",
      "92/92 [==============================] - 6s 70ms/step - loss: 1.5195e-06 - val_loss: 2.6067e-05\n",
      "Epoch 58/60\n",
      "92/92 [==============================] - 6s 69ms/step - loss: 1.6029e-06 - val_loss: 2.7867e-05\n",
      "Epoch 59/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.6748e-06 - val_loss: 2.1394e-04\n",
      "Epoch 60/60\n",
      "92/92 [==============================] - 6s 68ms/step - loss: 1.5303e-06 - val_loss: 2.5124e-05\n"
     ]
    }
   ],
   "source": [
    "# Train model\n",
    "history = model.fit(x_train, y_train, epochs=epochs, validation_split=validation_split, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c9ada8c8",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot loss against epochs to check for overfitting\n",
    "loss_hist = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.plot(loss_hist, label=\"Loss\")\n",
    "plt.plot(val_loss, label=\"Validation Loss\")\n",
    "plt.xlabel(\"Epochs\")\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c8b0a9b",
   "metadata": {},
   "source": [
    "## Loss v Epoch graph\n",
    "\n",
    "The validation loss does not show any upward trend as the model is trained so there are no signs of overfitting. However, validation loss didn't converge with the loss curve as I expected, there might be some underfitting."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "b7490723",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "44/44 [==============================] - 1s 22ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Test\n",
    "y_pred_scaled = model.predict(x_test)\n",
    "y_pred = prediction_scaler.inverse_transform(y_pred_scaled)\n",
    "y_test_unscaled = prediction_scaler.inverse_transform(y_test)\n",
    "\n",
    "# Plot actual vs predicted prices\n",
    "# Plotted price is the first day of the 10 days predicted (to act as a proxy for overall accuracy)\n",
    "plot_y_test = []\n",
    "plot_y_pred = []\n",
    "for i in range(y_test_unscaled.shape[0]):\n",
    "    plot_y_test.append(y_test_unscaled[i][0])\n",
    "    plot_y_pred.append(y_pred[i][0])\n",
    "\n",
    "plt.figure(figsize=(10, 5))\n",
    "plt.plot(plot_y_test, label='Actual Price')\n",
    "plt.plot(plot_y_pred, label='Predicted')\n",
    "plt.title(label=f\"Test dataset: NVDA stock price since {test_start_date}\")\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2467246f",
   "metadata": {},
   "source": [
    "## Reflection on model performance and settling on final model\n",
    "\n",
    "Before landing on my final prediction model, I tested different iterations of my model (e.g. differing hyperparamters, differing neural network structure). Across all these iterations, I found a unifying trend - the predicted prices were consistenly below the actual price of the stock (underpredicting). Furthermore, I noticed that the divergence worsened for the most recent dates. \n",
    "\n",
    "Eventually, I identified the main problems to be too many LSTM layers which were \"remembering\" too much historical data and pulling down the predicted prices. This phenomena was worse for recent dates (possibly) due to NVDA's bull run in 2023. Therefore, I limited my model to 1 LSTM layer and have seen improved results. \n",
    "\n",
    "## Hypothesis \n",
    "\n",
    "Based on the above observations, I hypothesised that the model was scaling down its predictions in some consistent manner. To test this, I plotted the error rate of the predictions to check if the error follows any discernable distribution.\n",
    "\n",
    "I also wanted to check if the errors were worse for the most recent trading days. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d6a5d4ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Find difference between predicted and actual prices in test data\n",
    "difference = y_pred - y_test_unscaled\n",
    "avg_diff = np.mean(difference, axis=1)\n",
    "y_test_avg = np.mean(y_test_unscaled, axis=1)\n",
    "\n",
    "#Convert price difference into percentage error\n",
    "error_perc = []\n",
    "for i in range(avg_diff.shape[0]):\n",
    "    error_perc.append(avg_diff[i] / y_test_avg[i])\n",
    "\n",
    "# Check whether percentage error roughly follows a normal distribution.\n",
    "plt.hist(error_perc, bins=len(error_perc))\n",
    "plt.title(\"Average percentage error in predicted prices\")\n",
    "plt.xlabel('Percentage error')\n",
    "plt.show()\n",
    "\n",
    "#I split the test data into two to check if the model underpredicted for later dates more than for earlier dates\n",
    "slice_test = int(np.ceil(0.8*y_test.shape[0]))\n",
    "error_perc_early = error_perc[0:slice_test]\n",
    "error_perc_recent = error_perc[slice_test:]\n",
    "barrier_date = data_filtered.index[split_data + slice_test]\n",
    "plt.hist(error_perc_early, bins=len(error_perc_early), label=f\"Error ({test_start_date} to {barrier_date})\", alpha=0.7)\n",
    "plt.hist(error_perc_recent, bins=len(error_perc_recent), label=f\"Error ({barrier_date} to present)\", alpha=0.4)\n",
    "plt.xlabel('Precentage error')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a676ba8",
   "metadata": {},
   "source": [
    "## Analysing histogram \n",
    "\n",
    "Throughout my testing, I found that the error rates on the predictions did in fact seem to follow a sort of normal distribution. However, in my final model, there was no significant bias towards underprediction (as seen in the first graph).\n",
    "\n",
    "\n",
    "Furthermore, as seen in the second graph, the distribution is not random - the predictions are more underpredicted for recent dates. I was initially going to adjust all my predicitions with the average error produced in the test dataset. However, I opted against it as it didn't seem to be a systematic method and the lack of random distribution would make the error correction inprecise. \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6e8cdc3f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scaled RMSE is 0.030902088093205316\n",
      "RMSE is 15.206144828436642\n"
     ]
    }
   ],
   "source": [
    "# Metric testing\n",
    "MSE_scaled = mean_squared_error(y_test, y_pred_scaled)\n",
    "RMSE_scaled = sqrt(MSE_scaled)\n",
    "print(f\"Scaled RMSE is {RMSE_scaled}\")\n",
    "\n",
    "\n",
    "MSE_unscaled = mean_squared_error(y_test_unscaled, y_pred)\n",
    "RMSE_unscaled = sqrt(MSE_unscaled)\n",
    "print(f\"RMSE is {RMSE_unscaled}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "50fb9276",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 16ms/step\n",
      "                 Close\n",
      "Date                  \n",
      "2023-10-24  414.418396\n",
      "2023-10-25  412.031219\n",
      "2023-10-26  408.684357\n",
      "2023-10-27  406.864441\n",
      "2023-10-30  413.043732\n",
      "2023-10-31  417.816376\n",
      "2023-11-01  422.995697\n",
      "2023-11-02  425.004822\n",
      "2023-11-03  418.336365\n",
      "2023-11-06  421.306885\n",
      "2023-11-07  435.494385\n"
     ]
    }
   ],
   "source": [
    "#Make predictions\n",
    "project_test_set = [np_data_scaled[len(np_data_scaled) - WINDOW:len(np_data_scaled), :]]\n",
    "project_test_set = np.array(project_test_set)\n",
    "project_predict_scaled = model.predict(project_test_set)\n",
    "final_predict = prediction_scaler.inverse_transform(project_predict_scaled)\n",
    "final_predict = np.transpose(final_predict)\n",
    "\n",
    "#Format predictions as DataFrame\n",
    "predicted_dates = ['2023-10-24','2023-10-25', '2023-10-26', '2023-10-27', '2023-10-30', '2023-10-31',\n",
    "                   '2023-11-01', '2023-11-02', '2023-11-03', '2023-11-06', '2023-11-07']\n",
    "\n",
    "df = pandas.DataFrame(final_predict, index=predicted_dates, columns=['Close'])\n",
    "df.index.name = 'Date'\n",
    "print(df)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "83fd19b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot predicted price with actual price since June 2023\n",
    "plot_start_date = '2023-06-01'\n",
    "visual_data = data_filtered.loc[plot_start_date:]\n",
    "plt.plot(visual_data.index.values, visual_data['Close'], label='Actual')\n",
    "plt.plot(df.index.values, df['Close'], label='Predicted')\n",
    "plt.title(label=f\"NVDA stock price since {plot_start_date}\")\n",
    "plt.legend()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea4f804d",
   "metadata": {},
   "source": [
    "## Final Thoughts\n",
    "\n",
    "Despite reducing underprediction in my model while testing, I still have the sense the latest iteration of my model is underpredicting prices. This is probably because 2023 has been an anamoulous year of trading for NVDA and my training dataset ends in 2018. Ideally, I would like to allow the model to train on data drawn from across the entire timeseries but due to the sequential nature of the data, I was not not able to scramble the data. \n",
    "\n",
    "I would also thinking about whether I would have been able to predict 10 days of trading prices sequentially rather than at one go as I suspect this might have been more accurate. However, this would only work if my input data was 2 dimensional (only using 'NVDA Close' price as a feature).\n",
    "\n",
    "Lastly, it would have served me well to automate testing with different hyperparameters and officially document the different tests I conducted rather than taking informal notes. \n",
    "\n",
    "Overall, I think the project helped me develop some intuition about how the different hyperparamaters and model architecture affect by predictions. I was driven to conduct data analysis on my predictions and I was able to successfully identify excessive LSTM layers as the root of the inaccuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "76dacb3d",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b7f87143",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}