Seeking Alpha

Contextuall's  Instablog

Contextuall
Send Message
Contextuall was founded on the belief that people’s online activities are a leading indicator of what they plan to do in the future. - Before you buy a car, you search for the best deals online - If you just lost your job, you tweet about it - If you’re visiting your local shopping mall, you... More
My company:
Contextuall
My blog:
Contextuall Free Reports
  • Regional Banking Stocks Ready For A Short Squeeze? Probably Not 0 comments
    Aug 1, 2012 10:25 AM | about stocks: KRE, ABCB, BBT, CFFI, FBNC, NKSH, UCBI, WSFS

    As we pointed out over the weekend, regional banking is one of the most shorted industries at the moment.

    To illustrate the average short float by industry:

    Consumer Services: Short float = 11.81 days of avg volume (7 companies)

    Regional Banks in Southwest: Short float = 11.68 days of avg volume (21 companies)

    Regional Banks in Southeast: Short float = 11.22 days of avg volume (27 companies)

    Companies Selling Drug Related Products: Short float = 11.01 days of avg volume (7 companies)

    Regional Banks in Pacific: Short float = 10.89 days of avg volume (42 companies)

    Northeast Regional Banks: Short float = 10.86 days of avg volume (66 companies)

    Movie Production & Movie Theaters: Short float = 10.68 days of avg volume (8 companies)

    Sporting Goods: Short float = 10.05 days of avg volume (6 companies)

    Medical Equipment Wholesale: Short float = 9.7 days of avg volume (6 companies)

    Industrial Equipment & Components: Short float = 9.59 days of avg volume (18 companies)

    Does that mean we can expect a short-covering rally anytime soon? Dick Bove seems to think so.

    It's crazy not to own the regional banks right now, he said in a CNBC interview this week.

    "The quality of earnings is extraordinarily high. It's coming from more loans and more deposits."

    That might be true eventually, but we don't expect this rebound in regional banking stocks to happen in the near term.

    Contextuall built 4 models to try and predict the direction of KRE, the SPDR KBW Regional Banking ETF (used as a proxy for the entire industry).

    Three of these models are predicting more losses for the industry over the next 10 trading sessions.

    To access the model descriptions and download backtests, simply click on any of these links.

    KRE Model 1 (Current Signal: None)

    Model Type: Linear Regression Model

    Based on an out of sample test between 3/1/2011 and 7/13/2012, this model achieved a directional accuracy rate of 65.82%. Based on the 7/27/2012 close, the model is predicting a KRE change of -2.45% over the next 10 trading sessions. Because the absolute value of this change is smaller than the threshold used in the backtest, no trading signal has been generated (i.e. the model does not have a definitive direction for the stock at the present time).

    KRE Model 2 (Current Signal: Sell)

    Model Type: Linear Regression Model

    Based on an out of sample test of Sell signals between 3/1/2011 and 7/13/2012, this model achieved a directional accuracy rate of 60.38%. After plugging in data on the KRE(n) close on 7/27/2012 (27.20), the model is projecting a KRE(n+10) value of 26.37, a threshold change of -3.06%.

    Because the absolute value of this projected negative change is larger than the threshold used in the out of sample test, a Sell signal has been generated (i.e. the model predicts that KRE(n+10) will be lower than KRE(n)).

    KRE Model 3 (Current Signal: Sell)

    Model Type: Linear Regression Model Based on an out of sample test of Sell signals between 2/24/2012 and 7/13/2012, this model achieved a directional accuracy rate of 58.54%.

    KRE Model 4 (Current Signal: Sell)

    Model Type: Bayesian Logistic Regression

    Based on 726 observations between 4/14/2008 and 2/28/2011 (the in-sample), Contextuall used a Bayesian Logistic Regression algorithm to predict if KRE will lose more than 2% at the end of the next 10 trading sessions. After testing the model's performance on an out of sample test based on 347 observations, the model correctly classified 238 historical instances, for an overall accuracy rate of 68.59%.

    Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

    Themes: Regional Banks Stocks: KRE, ABCB, BBT, CFFI, FBNC, NKSH, UCBI, WSFS
Back To Contextuall's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (0)
Track new comments
Be the first to comment
Full index of posts »
Latest Followers

StockTalks

  • $$ [new post] Intelligent Machines Reading the News, and Predicting the Future http://bit.ly/16juxxB
    Apr 30, 2013
  • $$ Just launched a graphical news sentiment tool for major companies--any thoughts? Please let us know :) http://bit.ly/ZlN89l
    Apr 23, 2013
  • $$ List of most bullish sectors and industries, based on aggregated tweets over last 24hours http://bit.ly/XIyyI0
    Apr 2, 2013
More »

Latest Comments


Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.