I will report here some mistakes incurred in the book and further explanatory comments. Apologies for the printed errors!

**Chapter 1**

p 34 1.3.4 RLab. There is an extra } in the first instruction. It should be

symbols=c('^VLIC','GE','KO','AAPL','MCD')

(See also the RLabs 1 in this web page for the corrected code)

**Chapter 2**

p 54 l4: "go back to Example 2.3" ==> "go back to R Example 2.3"

(or to be more specific "go back to Theorem 2.1 and the discussion that follows")

**Chapter 4.**

p 110: Final sentence of the preface to this chapter I promise to

"end the chapter with a case study of augmenting the set of information F_{t-1} with a sentiment measure drawn from a social networking platform.."

but I did not deliver such application at the end of the chapter. My apologies. It was a last minute decision to not include such application for reasons of space and time of delivering the book, and forgot to remove the aforementioned announcement. This was intended to be a summary of own work on using Twitter as a source of information for financial stocks prediction. The reader can alternatively read the original paper:

M. Arias, A. Arratia & R. Xuriguera, Forecasting with Twitter Data. ACM Transactions on Intelligent Systems and Technology, (TIST) vol. 5, n. 1, 2013.

(available at my web page --> Publications)

or see the following presentation which also contains R simulations of nonlinear models

p 121: line -8: the following math text is missing:

and

(This is not my fault for it does appears in my original files, so it got lost in the process of publication )

p 132 Fig. 4-4: The hidden layer of the nnet last hidden node should be

(not )

**Chapter 5**

p 152, l -12: "we obtain an approximation to the present value of the stock"

should say "we obtain an approximation to the value of the stock at time T"

(The fact that expected rate of return at 16% must be discounted -by multiplying the mean of by - is due to arbitrage: we have to pay for what we knew from the beginning will get for free. This is later clarify in sec 5.3 on Monte Carlo Methods).

p 155, l -5: sumarize --> sumarized

p 163, Fig 5.5.: Binomial Trees for American call options: I should have explained the numeric labels that appear in the tree (some readers have complaint to me about this). These corresponds to the value of the stock according to the binomial model minus strike K, i.e. the profit that one could get if the option is exercised at that time step. for example the top branch is S*u^i - K, with i=1,..,5 (minus C the price of the call)

p 167: Algorithm 5.2: Between instructions 2 and 3, include a resetting

of S = S_0 = 100, so that the new simulated path stars on initial value and not on the previously computed path's final value.

Also move in between those lines the sum=0 for similar reasons.

p 167 ff: rnorm(0,1) should be rnorm(mean=0, sd=1). If one doesn’t specify the parameters then the default of rnorm() is to consider first parameter as the size of the input vector, in this case it takes as 0 and the whole computation outputs 0.

**Chapter 6**

p 187-188: The algebraic equations describing the local extrema for the Head-and-Shoulders (HS) pattern are adapted from like equations that appeared in

Andrew W. Lo and Harry Mamaysky and Jiang Wang, *Foundations of Technical Analysis: Computational Algorithms, Statistical Inference, and Empirical Implementation, *The Journal of Finance, 55, 1705--1770, 2000.

This is a seminal paper in the subject of automatization of TA, one that I like very much, and had all the intention of including it as reference for this topic in the Bibliographic remarks (section 6.3.1), but had unexplainable forgotten. My apologies to the authors!

**Chapter 7**

p 226, l-3: remove one “Dissertation” (appears twice)

p227, l -5: A period (.) missing before “Then”

**Chapter 8**

p262: Cover Gluss (1986) —> Cover and Gluss (1986)