100 points by timeseries_forecasting 6 months ago flag hide 14 comments
curiousengineer 6 months ago next
I'm excited to see this Time-Series Forecasting System hiring Data Scientists! I've been trying to learn more about working in the field - can anyone share their experiences working on TSFS or similar projects?
experienceddatascientist 6 months ago next
I've had the pleasure of working on a few TSFS-like projects, and I can confirm, they're challenging yet immensely rewarding. Excited for the opportunity of this YC S21 team!
curiousengineer 6 months ago next
That's great to hear! I'm particularly interested in the applications of such systems. Any chance you could share where TSFS-like projects have been useful in your experience?
experienceddatascientist 6 months ago next
Certainly! I've worked on projects including stock market prediction, weather forecasting, and even trend analysis for social media platforms. TSFS is a potent tool.
newtodata 6 months ago prev next
TSFS sounds fascinating. Could someone explain the key differences between Time-Series Forecasting and traditional Machine Learning forecasting techniques?
mlphd 6 months ago next
Time-Series Forecasting focuses on modeling and predicting data points indexed in time order, enabling the analysis of trends, seasonality, and other dynamic factors. ML techniques are versatile and not solely time-dependent, making them better suited for broader classification tasks.
newtodata 6 months ago next
Thanks for explaining. This specialization seems to open several doors. Have any resources to share that would help an aspiring Data Scientist to focus on Time-Series Forecasting techniques?
mlphd 6 months ago next
I highly recommend the book 'Forecasting: Principles and Practice' by Athanasopoulos and Hyndman. It provides a strong foundation in the principles and various TS methods, including ARIMA and state-space modeling.
learningfanatic 6 months ago prev next
Any particular programming languages or libraries popular in TSFS development? I'm keen to begin developing my skills in this domain.
statsguru 6 months ago next
Python is the most popular language nowadays for TS projects, with libraries like Statsmodels, Prophet, and Tensorflow offering robust time-series functionality.
learningfanatic 6 months ago next
Thanks! I'm an avid R user, but I think I'll give Python and those libraries a go toexperiment with their TS functionality.
hiringengineer 6 months ago prev next
@curiousEngineer, we're looking for Data Scientists skilled in ARIMA, ETS, Prophet, and deep learning techniques to join our Time-Series Forecasting System team. If this meets your expertise and interests, DM me!
curiousengineer 6 months ago next
@hiringEngineer, thanks for sharing. I'm very DMing you. :)
skilledmindset 6 months ago prev next
Impressive to see the innovation in Time-Series Forecasting! Between ML and TSFS, what would you consider the future for forecasting methods, fellow HNers?