DataDreams - A blog for data science and machine learning
Story
bull stock market data science
Date
Sep, 28 2018
Stocks/Securities
Talking About

A Decade for the Bulls! Visualizing 10 years of Stocks, Commodities, and Currencies

Since the 2008 recession, stocks have been on the longest bull run in U.S. history, carried mainly by the technology sector.

2018 has been interesting on a macroeconomic scale: the U.S. Dollar Index has strengthened, the Argentine Peso has plummeted, Crude oil is in the middle of a resurgence, the yield curve has narrowed, and the U.S. stock market has strengthened in relation to intern...

Story
crypto-trading
Date
Sep, 01 2018
Machine Learning/AI
Talking About

Using a Keras LSTM to predict Cryptocurrency prices

LSTM's, or Long short-term memory units, are a type of recurrent neural network that are useful in processing sequential data. They keep their hidden state over time, and so have a form of memory.

If you learn one thing from this post, it is that a long short term memory network as simple as the one in this article is not that great at predicting something as random as the swings, and often t...

Story
time series forecast
Date
Aug, 24 2018
Machine Learning/AI
Talking About

Preparing data using Pandas for time series forecasting

If you are interested in predicting the future movements of a dataset, you are going to need to setup the data so that it conforms to a certain format for time series training.

This article is not going to be a thorough treatment about time series analysis, so concepts like seasonality and other factors that might affect the accuracy of time series prediction are not in the purview of this tut...

Story
kdb-database
Date
Aug, 01 2018
Big Data
Talking About

A tutorial with kdb+: the hidden secret for big data at speed

This will be a quick lesson on downloading and running a few queries in kdb+, a lightweight (seriously the binary download is 300kb) database that is used in a variety of applications where performance for vast amounts of data is critical. From quant firms trading financial securities to F1 teams needing streaming telemetry data for their race cars, kdb+ is used in a variety of highly demanding real ...

Story
banks and the one year gap ratio
Date
Jul, 26 2018
Stocks/Securities
Talking About

Banking winners and losers in a rising interest rate environment

In any given year, there will be assets that are set to reprice, and there are liabilities set to reprice. The difference between these two amounts as a percentage of total assets is called the one-year gap ratio.

This ratio is particularly interesting for investors studying banks in a rising interest rate environment like the one we are currently experiencing in 2018, with the fed signaling ...

Story
nlp and oscar wilde
Date
Jul, 23 2018
NLP
Talking About

Spot the author: identifying famous authors with just a small list of words

In this post, I'm going to post five lists of 50 or so words obtained using natural language processing. These lists are words from select works by famous novelists.

Sometimes the list came from one seminal work, and other times the list was generated from multiple popular works from the mystery authors in question. I was particularly interested in words that the author used in more than on...

Story
anscombes quartet
Date
Jul, 21 2018
Data Science
Talking About

Anscombe's Quartet: why you should visualize your data

If you looked at the graphs on the image to the left, you would think that the underlying data points wouldn't share much in common. However, all three data groups share the same mean value. That's right, the average value for these three disparate visual representations are all the same. Not only that, but they also have the same Pearson correlation coefficient.

Pretty neat, right? Well, Fra...