Lumengram Posted July 28, 2019 Report Share Posted July 28, 2019 As the price of an uber ride surges in different hours, I feel that it can be modeled using a cos or sin function. Do you think this topic will be sufficient for a good IA? Thank you! Reply Link to post Share on other sites More sharing options...
kw0573 Posted July 28, 2019 Report Share Posted July 28, 2019 Theoretically speaking, the topic is ok. But data may not be what you expect and a sinusoidal function might not adequately model the price fluctuation. Reply Link to post Share on other sites More sharing options...
Lumengram Posted July 28, 2019 Author Report Share Posted July 28, 2019 Thanks for the reply! Assuming that the price peaks twice a day(once in the morning, and one in the evening), what kind of function would be appropriate? In the scenario that the pattern is more complex in which there are also smaller price fluctuations throughout the day, what would you suggest I do? I was told that a Fourier series or a Taylor series would be more suitable for this kind of modeling, but would involve more advanced math. Reply Link to post Share on other sites More sharing options...
kw0573 Posted July 28, 2019 Report Share Posted July 28, 2019 Obviously you need the data first to know what the function is. Once you plot the price against time then it should be clear what functions it is. Reply Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.