Mortgage Refi Index Prediciton within 4.6%

The MBA posted their weekly numbers on Wednesday, with refinance applications coming in at an index level of 1862. This number falls within 4.6% of the Markets with Search prediction of 1950. This gap is slightly larger than would have been anticipated. Looking to this predictive set of keywords in the future, we will be adding a more comprehensive set to provide more accurate predictions.

Tracking Mortgage Applications with Search

Weekly data releases tend to have a little bit less pull in markets than monthly releases, but they can still have a strong impact. One of the strongest weekly indicators that we see released each Wednesday is the Mortgage Bankers Association Mortgage Applications Indexes.

The data released on Wednesdays comes from the prior week ending Friday. While a 5 day lag may not seem like a long time, we can narrow that number by using the power of search data. Taking a look at queries on Google for “mortgage refi” and “mortgage refinance,” it turns out that these indications of interest correspond very highly to action and applications. Comparing these search indexes with the MBA’s Refinance Index, there is an R² of 0.8121.

The last release of the Refi Index came in at 2089.7 for the week ending July 17^th. Data for the week ending July 24^th will be released this Wednesday, July 29^th. Looking at search data, there has been an 8.7% decrease in popularity in searches for this week. This should yield a similar movement in the Refi Index.

Assuming that the Refi Index would follow in a similar relativity to search, this week’s index should come in at ~1950.

Bank of Israel Using Search Data to Get a Pulse on Economic Activity

The Bank of Israel recently posted a brief analysis highlighting how they have looked at Google Insights for Search to predict economic activity. I highly encourage reading through the white paper linked from the press release; it truly shows the power and depth you can get into when analyzing search data.

A very interesting result from this analysis shows some good news for the Israeli economy:

In the last few months the query indices in Israel point to a reduction in the probability of a slowdown; that conclusion is based on the moderation of the decline in private consumption. The indices do, however, indicate a steep increase in unemployment in the second quarter of the year, to more than 8 percent.

It is reassuring to see these kinds of analyses pop up outside of Markets with Search. Thanks for the tip on this press release, Hal.

I will try to repost these articles as I come across them to point to ideas outside of MWS. Have a great weekend and happy trading.

Building with Housing Starts Data

Residential real estate has been at the heart of market movements since the subprime crash dating back to February 2007. Housing inventory has been skyrocketing with foreclosures getting to all time high levels. This has put significant pressure on the home building sector.

Home building has not ceased, but has significantly decreased. New housing starts in June 2009 were at a level 28% of where they were at the same time in 2005. These starts have ramifications outside of just homebuilders like DR Horton (NYSE:DHI), Pulte Homes (NYSE:PHM), and Lennar (NYSE:LEN). Suppliers for new home supplies are impacted as well.

Searches for key indications of interest in the housing sector fall very closely in line with where the true number falls. Going back to 2005, there is an R² of 0.7667 with searches for these key indicators and the true housing starts number.

Markets with Search (tentative) prediction for July 2009 housing starts: flat. The numbers will be out in mid-August.

Updated Apple Earnings: iPhone Sales Prediction within 100k

Apple just released their earnings for fiscal Q3 2009. Consensus picks for iPhone units sold came in between 4.5 and 5.0 million; the Markets with Search prediction came in at 5.3 million.

The final number of quarterly sales: 5.2 million. While the data points were not large to choose from, the lift in search traffic turned out to be an accurate predictor of this data indicator.

Hopefully more of these predictions will reach the same accuracy in the future.

Ahead of Earnings: Predicting iPhone Sales

Apple Inc (NASDAQ:AAPL) is set to release earnings for their fiscal Q3 2009 on Tuesday, July 21. A large part of the speculation into the shareholder value of Apple is the success of their iPhone, which launched on June 19^th. In fiscal Q2 2009, Apple sold 3.79 million iPhone units globally.

Predicting the performance of iPhone sales can be tough if looking simply at searches for the term “iPhone.” These searches are subject to the volatility of news, branding, and buzz around the product. An easy filter to take a peek into how sales are performing is to add call to action keywords into search terms. Since the product is global, the analysis below takes a global look at “buy iphone” searches in English, Spanish, Russian, Japanese, and German (Apple’s top iPhone market languages).

While there are not yet enough data points to run a statistically significant regression on searches to sales, eyeballing the data looks quite astounding.

The major global launch of the iPhone 3G in July 2008 appears to have skewed the data for calendar Q2 and Q3 2008, yet all other search movements appear to proportionally fall and predict the according iPhone sales numbers for that quarter. Taking Q2 and Q3 as outliers of the globalization of the iPhone, the most recent search numbers for the past quarter’s lift would indicate that iPhone sales will rise 40% in the earnings release on Tuesday.

Markets with Search prediction for fiscal Q3 2009 Apple iPhone unit sales: 5.3 million.

Predicting the Present

The coming posts to Markets with Search will provide further analysis similar to the jobless trends data. These analyses aim to provide actionable data-oriented plays on the market. Before getting into more, it is important to note some other sources that are out there for your reading. I’ll be posting these kinds of articles intermittently.

The Chief Economist of Google, Hal Varian, has published “Predicting the Present with Google Trends.” This whitepaper embodies the power of the real-time nature of internet data; we can increase transparency into lagged data. In this post, Varian explores how to predict auto sales, home sales, retail sales, and travel behavior using the wisdom of the online crowd.

All of these factors can issue trade ideas. Play around with your ideas on Google Insights or Microsoft xRank.

Enjoy the light weekend read.

Getting Ahead of Initial Jobless Claims

Have you ever wanted to get ahead of data before the official release to the street? Have a play based on the numbers?

The time has finally come. After doing some backtesting, some content has been geared up and ready for release. Today’s trend: using Google to predict where seasonally adjusted job claims will turn… with 88% confidence.

Consumers turn to the search engine juggernaut to indicate interest in topics that have been on their minds, making it the world’s largest focus group. One of the most interested topics of 2009: What do I do if I just lost my job?

Answer: search for filing unemployment. With an R² of 0.8815, searches for “file unemployment” and “filing unemployment” fall right in line with actual seasonally adjusted Initial Jobless claims for a given week.

It is important to note that the indexing algorithm for Insights for Search readjusts with every new peak in popularity, requiring you to index SA Initial Claims accordingly using the methodology (SA Initial Claims_n / Max SA Initial Claims * 100). Given the values from July 16, 2009 the following model will allow you to fairly accurately predict the movement in jobless claims:

Predicted SA Initial Claims = 5006.2 * (Search Index for Unemployment Filing) + 225,046