{
  "fetched_at": "2026-10-20T05:48:42Z",
  "source_url": "https://www.kaggle.com/competitions/econometrics-2-w20/data",
  "pages": [
    {
      "name": "data-description",
      "content": "# File descriptions\n- `data.rds` - A prepared **R** object with the training and test datasets.\n- `train.csv` - The training set with the dependent `price` variable.\n- `test.csv` - The test set with all variables but `price`.\n- `sample_submission.csv` - A sample submission file in the correct format.\n- `variable_description.md` - A text file describing each variable in the dataset.\n\n# Usage\n- You can read in the prepared *RDS* file in **R** using: `x <- readRDS(\"data.rds\")`.\n- The *CSV* files can be read using: `train <- read.csv(\"train.csv\")`.\n- A submission can be created with: `write.csv(prediction, \"submission.csv\", row.names = FALSE)`.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Welcome to this Kaggle InClass competition as part of the \"Econometrics 2\" course (number 2246) at WU Vienna.\n\nIn this competition you are provided with a dataset on house prices and potential explanatory variables. This dataset is split in a *training* subset with and a *testing* subset without house prices. Your goal is to predict these missing prices using the data provided to you.\n\nThe data has been cleaned and prepared for you, so you can focus on data transformation, model specification, model selection, and actual estimation. There is a relatively large number of potential explanatory variables, so be wary of overfitting and poor out-of-sample performance.\n\nYour predictions are evaluated in two steps. During the competition, your submissions receive a preliminary score based on a part of the testing dataset, which will appear on the public leaderboard. Your final score is based on the other part of the testing dataset and will only be visible once the competition has finished - making good out-of-sample performance crucial. As a result, your final score and rank may change considerably from the public leaderboard. \n\n### Acknowledgements\nI thank the creators and provider of the dataset and Prof. Geyer for the idea and his highly valued inputs.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "The evaluation metric for this competition is the root mean squared logarithmic error based on (natural) logs of actual and predicted values:\n$$RMSLE = \\sqrt{\\frac{1}{n}\\sum_{i=1}^{n} (\\ln y\\_i - \\ln \\hat{y}\\_i)^2}.$$\n\nPublic and private scores are calculated from different subsets of the dataset. Your final rank, score and position on the private leaderboard will only be available after the competition has closed.\n\nThe RMSLE of two values can be calculated in **R** using the following function:\n```\nrmsle <- function(true, fitted) {sqrt(mean((log(true) - log(fitted)) ^ 2))}\n```\n\nSubmissions must be made as *CSV* (comma-separated-value) files containing two columns -- the `id` and the predicted `price`. See the data page for an example file.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "- One account per participant\n- No sharing code outside of teams\n- Submissions must be reproducible\n- Only use data or information that is provided to you\n- Refrain from using black-box models",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}