Machine Learning Algorithm for Linear Nonparametric Forecasting

Experienced data scientist who has worked on projects ranging from recommending engines to computer vision applications. Have extensively used all kinds of machine learning algorithms like decision tree based models, linear models & neural networks. Would like to know more about the dataset. Looking forward to hearing back from you.

Hi... I am a Python and Machine Learning specialist, certified by Freelancer. I fully understand your project and I am sure I can help you. Let's discuss details by chat.

Hello, Do you also have patient data? I believe patient profiling will lead us to more realistic prediction. Your problem can be viewed as something like product recommendation, where a particular type of patient is likely to buy again or not. Patient data like: gender, age, income etc. should be good enough to start with. Regards, Samiran

Hello I can achieve this project perfectly using R I master machine learning algorithm , statistics ,time series ... please contact me for more details about the project best regards

engineer in statistics and economics applied and technician in microsoft office. I am ready and able to do this job

Hello! I'm glad to see a good description. Good work. So, I wanted to point out a few things that seem wrong to me, about your data structure or fields. 1) DateOfPicked, DateOfPrescribed, DateOfExpiry/Validity 2) DaysAfterPicked, DaysValid You can keep either of the formats above. Your current format doesn't give any information about how many days after it was picked. Am I right? About project So, we can find the likelihood of picking up the drugs on a certain day (of 30 days validity) So, each patient will have a 0 to 1 real valued score for all the days of validity. To make it a classification, whether it is picked ON TIME/NOT, we can integrate the probabilities and assign 0 or 1 as is. To compare 2 patients, same numbers can be used. You could answer like: What's the likelihood the P1 picks up by 7th day or P1 picks up 7 days earlier than P2? Machine learning/Platform This is a regression problem, not classification. I can use R or Python. Pick your preference. I'll pick Python 2.7. Supervised learning So, first we need to process the data to find the likelihood values. And, then regression analysis CART for finding the function Queries Why do you call it non-parametric? In DATA driven model, it is non parametric at the beginning, predictor doesn't have any equation apriori, but at the end, it is parameters weights only. Anyway, it's not relevant here. Thanks! Fields Add more columns and features if you can About me I work as research assistant at computer vision and machine learning lab in my university. Thank you!

Hello, I have extensive experience in python and financial markets and would love to help you out on this project. I currently work part-time as a consultant for a long/short equity hedge fund part/time in Palo Alto, CA and have a Masters in financial mathematics/engineering. I think I can be of great help to you. My credentials are below: * Responsible for creating proprietary quantitative models and algorithmic trading strategies for long/short equity optimization models with specific risk and return parameters specified by the investor profile, utilizing machine/deep learning, along with Q reinforcement learning agents. •Created predictive vectorized/event-based machine learning models utilizing multivariate/logistic regression, lasso/ridge regression, linear/quadratic discriminant analysis, decision trees, K neighbors, Naive Bayes, random forest, support vector machine, Adaptiveboost, GradientBoost, XGB, and portfolio optimization to maximize return and minimize volatility for various investor risk profiles. •Responsible for deep learning modeling using recurrent neural networks, Tensorflow, nltk, sentiment analyzer, Keras LSTM, and convolutional neural networks, in attempt to predict specific asset class forecasted prices through stocks, forex, bonds, futures, ETFs, and other derivatives. •Designed a proprietary machine/deep learning long/short intraday algorithm utilizing the Interactive Brokers API, IB_Insync python library. Speak soon!

Why don't you try WEKA for this task? I can provide Python and C++ code or WEKA model which is able to make these predictions.

We will develop the programme in tensor flow and use a back propogation neural net for accurate prediction the extent of training depends upon the size of the data so we will need access to the file to determin extent of training involved

Hi, I am full stack data scientist/engineer and I can help in your project. I can develop a machine learning model in python using sci-kit learn machine library. If you are not available now we can agree on timing to chat and discuss on the project. Please let me know about time when you would be available.

Currently working as a Quant Analyst/ Data Scientist at Gartner for their Talent Assessment Division. Major part of my job is to work with IO Psychologists to create top grade Talent Assessments software to identify best talents across job functions and geography. I create predictive tools to zero in on talents which have the best culture fit according to the clients. My daily drivers are R, Python, SPSS, Postgresql, Mysql, Excel, R shiny etc. Proficient in ggplot, scikit learn, numpy, scipy, Shiny, linear regression, decision trees. Contact me for more information.