 Basis $\alpha$ of V such that it is simultaneously orthonormal wrt $\langle,\rangle$ and orthogonal wrt $\phi$ bilinear form
 Bundle map inducing isomorphism on fibers  when is it an isomorphism?
 Does Kunneth formula work for sheaf cohomology with coefficients on the tangent bundle?
 Fourier transform of $e^{i\log(x)}$?
 Apply Leibniz integral rule to derive the formula for power $P=\mathbf{F}\cdot\mathbf{v}$
 About three conjectures concerning the Pythagorean Theorem.
 A problem regarding Chinese Remainder Theorem
 inner product space: closest point in a subspace is orthoganal
 Intuition behind the 2D heat equation and examining numerical solutions through inspection
 How can I prove that is A is a m by n matrix and m
 Prove big O for a logarithmic function
 Delta Epsilon Proof
 Give an example of a measure such that sine preserves this measure
 Probability density function of the bounded sum of independent random variables
 Question on Mathematical Logic & Book Recommendation
 Understanding the $\Delta$complex structure of a quotient space
 Prove or give a counter example of the following statement
 What is $\frac{1}{\omega_n}\displaystyle\int_{\mathbb{S}^n}\! x_{n+1}^2\,\mathrm{d}x$?
 Find pdf given transformation
 Show that for any sets A and B that A Δ B = A^c Δ B^c.
How to calculate “Revenue of next order per customer”?
I have transactional data of an online store, how can I predict the "revenue of next order per customer" or "time until next order per customer".
I have following columns (Invoicenumber,StockCode,Description,Quantity,InvoiceDate,UnitPrice,CustomerID,Country.)
That may or may not be enough data for truly predictive modeling. You have order information, but you don't have customer information.
You might be able to predict things like how many units of product X you are going to sell next you or the average shipping costs of orders for next year. But you won't be able to relate things on an individual customer basis (like your question says) until you have a lot more customer data.

That may or may not be enough data for truly predictive modeling. You have order information, but you don't have customer information.
You might be able to predict things like how many units of product X you are going to sell next you or the average shipping costs of orders for next year. But you won't be able to relate things on an individual customer basis (like your question says) until you have a lot more customer data.
20170321 00:38:29