The solution is 3 months ago, so if there is anything missing, please let me know.
\nThe objective of the competition is to predict monthly microbusiness density in 3315 cfips (County FIPS) in the United States. The code implementation incorporates several special features and techniques to improve the accuracy of the predictions.
\nThis approach predicts the microbusiness density for each of the 3315 cfips individually. This allows for more granular and accurate predictions tailored to specific regions.
\nThe code utilizes popular technical indicators such as Exponential Moving Average (EMA), Momentum (MOM), and Relative Strength Index (RSI). These indicators capture underlying trends, momentum, and market conditions, enhancing the predictive power of the model.
\nIn addition to the microbusiness density data, the model incorporates several external datasets. These datasets include information such as unemployment data, earnings, rent, DSG10, tax rate, housing price, and population estimates. By incorporating these relevant external factors, the model can capture the influence of broader economic and demographic factors on microbusiness density.
\nOptuna is used to minimize the SMAPE in the Catboost model.
\nThe code employs the Catboost model. Also used Cross Validation (CV) to generalize well to unseen data.
\nTo predict multiple future time periods, five separate models are trained. Each model is designed to predict a specific time horizon, such as t+1 month, t+2 months, and so on.
\nThe code \"6-external-datasets.ipynb\" is found on Kaggle and is provided to incorporate additional external datasets into the prediction model.
\n6-external-datasets.ipynb: add additional external datasets to the prediction model. It can be found on Kaggle and is intended to enhance the feature set.
kaggle_competition_microbusiness.ipynb: contains the full model training code for microbusiness density prediction. It encompasses data preprocessing, model training using Catboost and Optuna, and evaluation using SMAPE.
Github link: https://github.com/ttterence927/kaggle_competition_microbusiness/
\nIf you find this code implementation valuable or interesting, please consider giving it a star on GitHub.
", "comments": [ { "id": 2313153, "author_name": "@kaggleqrdl", "author_url": "", "post_date": "2023-06-22T13:29:32.683000", "content": "Well done on your solo gold. You can link to this from the leaderboard by updating the team page.
\nhttps://www.kaggle.com/competitions/godaddy-microbusiness-density-forecasting/team
", "votes": 1, "replies": [ { "id": 2313174, "author_name": "Terence", "author_url": "", "post_date": "2023-06-22T14:00:40.233000", "content": "Thank you :)) Learned a lot from you and all the people here. And congrats on winning 1st place!!
", "votes": 2, "replies": [ { "id": 2317208, "author_name": "ls", "author_url": "", "post_date": "2023-06-25T14:03:58.510000", "content": "impressive and congrats
", "votes": 0, "replies": [] } ] } ] } ], "comment_count": 3, "messages": [ { "id": 2313124, "postDate": "2023-06-22T12:52:01.667Z", "content": "The solution is 3 months ago, so if there is anything missing, please let me know.
\nThe objective of the competition is to predict monthly microbusiness density in 3315 cfips (County FIPS) in the United States. The code implementation incorporates several special features and techniques to improve the accuracy of the predictions.
\nThis approach predicts the microbusiness density for each of the 3315 cfips individually. This allows for more granular and accurate predictions tailored to specific regions.
\nThe code utilizes popular technical indicators such as Exponential Moving Average (EMA), Momentum (MOM), and Relative Strength Index (RSI). These indicators capture underlying trends, momentum, and market conditions, enhancing the predictive power of the model.
\nIn addition to the microbusiness density data, the model incorporates several external datasets. These datasets include information such as unemployment data, earnings, rent, DSG10, tax rate, housing price, and population estimates. By incorporating these relevant external factors, the model can capture the influence of broader economic and demographic factors on microbusiness density.
\nOptuna is used to minimize the SMAPE in the Catboost model.
\nThe code employs the Catboost model. Also used Cross Validation (CV) to generalize well to unseen data.
\nTo predict multiple future time periods, five separate models are trained. Each model is designed to predict a specific time horizon, such as t+1 month, t+2 months, and so on.
\nThe code \"6-external-datasets.ipynb\" is found on Kaggle and is provided to incorporate additional external datasets into the prediction model.
\n6-external-datasets.ipynb: add additional external datasets to the prediction model. It can be found on Kaggle and is intended to enhance the feature set.
kaggle_competition_microbusiness.ipynb: contains the full model training code for microbusiness density prediction. It encompasses data preprocessing, model training using Catboost and Optuna, and evaluation using SMAPE.
Github link: https://github.com/ttterence927/kaggle_competition_microbusiness/
\nIf you find this code implementation valuable or interesting, please consider giving it a star on GitHub.
", "rawMarkdown": "# **GoDaddy - Microbusiness Density Forecasting**\n\nThe solution is 3 months ago, so if there is anything missing, please let me know.\n\nThe objective of the competition is to predict monthly **microbusiness density** in **3315 cfips** (County FIPS) in the United States. The code implementation incorporates several special features and techniques to improve the accuracy of the predictions.\n\n## **Approach**\n\n### **1. Predicting 3315 Time Series**\nThis approach predicts the microbusiness density for each of the **3315 cfips** individually. This allows for more granular and accurate predictions tailored to specific regions.\n\n### **2. Technical Indicators**\nThe code utilizes popular technical indicators such as **Exponential Moving Average (EMA)**, **Momentum (MOM)**, and **Relative Strength Index (RSI)**. These indicators capture underlying trends, momentum, and market conditions, enhancing the predictive power of the model.\n\n### **3. External Datasets**\nIn addition to the microbusiness density data, the model incorporates several external datasets. These datasets include information such as **unemployment data**, **earnings**, **rent**, **DSG10**, **tax rate**, **housing price**, and **population estimates**. By incorporating these relevant external factors, the model can capture the influence of broader economic and demographic factors on microbusiness density.\n\n### **4. Optuna for Hyperparameter Optimization**\n**Optuna** is used to minimize the **SMAPE** in the **Catboost** model.\n\n### **5. Catboost Model with Cross Validation**\nThe code employs the **Catboost model**. Also used **Cross Validation (CV)** to generalize well to unseen data.\n\n### **6. Multiple Model Training**\nTo predict multiple future time periods, five separate models are trained. Each model is designed to predict a specific time horizon, such as **t+1 month**, **t+2 months**, and so on.\n\n### **7. External Dataset Addition**\nThe code \"**6-external-datasets.ipynb**\" is found on Kaggle and is provided to incorporate additional external datasets into the prediction model.\n\n## **Limitations**\n\n- **Limited Historical Data**: The current approach utilizes only the past three months of data to predict future microbusiness density. Exploring longer intervals of historical data could provide additional context and potentially improve the forecasting accuracy.\n\n## **Code Files**\n\n- **6-external-datasets.ipynb**: add additional external datasets to the prediction model. It can be found on Kaggle and is intended to enhance the feature set.\n\n- **kaggle_competition_microbusiness.ipynb**: contains the full model training code for microbusiness density prediction. It encompasses data preprocessing, model training using **Catboost** and **Optuna**, and evaluation using **SMAPE**.\n\n**Github link**: [https://github.com/ttterence927/kaggle_competition_microbusiness/](https://github.com/ttterence927/kaggle_competition_microbusiness/)\n\n## **Show Your Support**\n\nIf you find this code implementation valuable or interesting, please consider giving it a star on **GitHub**.", "votes": 8 }, { "id": 2313153, "postDate": "2023-06-22T13:29:32.683Z", "content": "Well done on your solo gold. You can link to this from the leaderboard by updating the team page.
\nhttps://www.kaggle.com/competitions/godaddy-microbusiness-density-forecasting/team
", "rawMarkdown": "Well done on your solo gold. You can link to this from the leaderboard by updating the team page. \n\nhttps://www.kaggle.com/competitions/godaddy-microbusiness-density-forecasting/team", "votes": 1, "replies": [ { "id": 2313174, "postDate": "2023-06-22T14:00:40.233Z", "content": "Thank you :)) Learned a lot from you and all the people here. And congrats on winning 1st place!!
", "rawMarkdown": "Thank you :)) Learned a lot from you and all the people here. And congrats on winning 1st place!!", "votes": 2, "replies": [ { "id": 2317208, "postDate": "2023-06-25T14:03:58.510Z", "content": "impressive and congrats
", "rawMarkdown": "impressive and congrats" } ] } ] } ], "complete": true, "source": "live", "fetched_at": "2026-10-19T19:30:19Z", "competition_id": 41881, "competition_slug": "godaddy-microbusiness-density-forecasting", "team_id": 9953701, "solution_url": "/competitions/godaddy-microbusiness-density-forecasting/writeups/king-hang-siu-public-531st-private-10th-the-comple" }