{
  "fetched_at": "2026-10-20T07:07:13Z",
  "source_url": "https://www.kaggle.com/competitions/gdp-forecast-using-data/data",
  "pages": [
    {
      "name": "data-description",
      "content": "The dataset comprises various variables, with detailed descriptions provided in the \"data summary\" sheet. The required files for analysis are:\n##Files\n*   **train.csv:** Contains the training data in CSV format.\n*   **variable_names.csv:** Lists the names and descriptions of the variables/Supplementary information regarding the dataset.\n*   **sample_submission.csv:** A template for submitting predictions in the correct format.\n\n**Format:** The variables are represented as integers or floats.\n\n**Target Variable:** The target variable for prediction is \"gdp_uae_18,\" which denotes the GDP for the United Arab Emirates (UAE). Acronyms used are explained in the variable summary sheet.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "The data collected for the challenge is via various sources. Your main objective for the challenge would be to utilize the features and build a data science model to forecast the GDP.\nYou will have independent variables to train on in the training set but you will have none of the independent variables in the test set as the future values of these are unknown.\n\n## Guideline\nThe following guidelines must be followed\n-Separate sections in the code for **EDA, Feature Selection, Missing Value Imputation, Model Development, Scenario Drivers and Results**\n\n- The variable `gdp_uae_18` should be the target variables \\. \n\n\n\n## Here the MAE is calculated using the following formula:\n\n$$ MAE =\\dfrac{ \\sum|{y_i-x_i}|}{n} $$\n\nWhere,\n\nMAE=\tmean absolute error\nyi\t=\tprediction \nxi\t=\ttrue value\nn\t=\ttotal number of data points\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "Submissions are evaluated on MAE (https://en.wikipedia.org/wiki/Mean_absolute_error) between the predicted variable and the observed target.\n\nThe submissions will also be evaluated based on the techniques used in the notebook to see how the missing values were filled and the scenarios were created.\n\nIn the first stage, the models with the lowest MAEs will be used for screening, then the models with the best scenario drivers and missing value techniques will be deemed as the winner.\n\n\n## Submission File\nFor each year in the test set, you must predict a forecasted value for gdp_uae_18. The file should contain a header and have the following format:\n\n    year,gdp_uae_18\n    2017,24560\n    2018,25555\n    2019,24321\n    2020,26789\n    2021,24352\n    etc.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "- Don't cheat!\n- Apply yourself!\n- Have fun!",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "abstract",
      "content": "The aim of this competition is to assess the capabilities of talented individuals proficient in data science, econometrics, and time-series analysis. Participants are tasked with constructing a Python-based model capable of forecasting GDP five years into the future. Evaluation criteria for determining the winner include:\n\n1. Minimization of Mean Absolute Error (MAE).\n2. Development of scenarios illustrating the effects of changes in variables like CPI, Unemployment,     Imports, or Exports.\n3. Imputation of missing values within the dataset.\n\n**Challenges:** The dataset covers the years 1975 to 2016 and contains numerous missing values across various variables. Participants must address these missing values and employ machine learning techniques to identify key features for GDP prediction.\n\n**Goal:** The primary objective is to predict GDP for the years 2017 to 2021 using a range of macroeconomic indicators. Additionally, participants are encouraged to explore scenarios demonstrating how alterations in factors such as CPI, Unemployment, Imports, or Exports influence GDP fluctuations. Efforts should also be made to address missing data within the dataset.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}