We are presenting this information to you so that you may

• Have a fundamental understanding of Financial Networks and their properties.

• Identify Systemically Important Financial Institutions.

• Add Financial Network Analysis to your investment analysis framework.

Financial networks are a map of which bank has lent money to some other bank. Analyzing the network can help identify banks that pose inherent risks to the system.

• Guide overall portfolio strategy
• Inform clients of potential pitfalls
• Identify specific bank stocks to short

• Density - ie how densely connected are the banks? Pre-Lehman = very dense
• Clustering (transitivity) - how interwoven groups in the network are
• Path length - like how many people separate you from another on linked in
• Degree - the total number of edges (transactions in this case)

Here is a nice orderly example of a network so we can have an understanding for a general model.

set.seed(7) 
e <- erdos.renyi.game(100, 0.1) #100 Nodes, Prob edge btwen two vertices = .10
plot(e)

Easy to calculate these in R;

graph.density(e) # Ratio of no. of edges vs no. possible edges

## [1] 0.1052525

transitivity(e) # Prob that adjacent vertices are connected

## [1] 0.09962687

average.path.length(e) # Average dist between vertices

## [1] 2.192929

1 There are some banks that are so important, that if they were to disappear the market itself would suffer greatly.

2 We can quantify that risk by defining "centrality"

• Degree : The total number of transactions a bank was involved in. The more transactions, the higher the degree and therefore the more important the bank is in the network.
• Betweenness : The number of times a node acts as a bridge along the shortest path between two other nodes
• Closeness : Describes the distance of separation
• Eigenvector : Measures the influence of a node in the network

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## 0.03856 0.11120 0.52070 1.00000 1.24300 4.65000

These are the banks (by number) that we will watch for downgrades by a debt rating agency due to rising leverage ratios.

names(which(index>3.5))

## [1] "18" "4"  "9"

When these banks reach the critical threshold in our model, we short the financials sector using XLF. library(c( "stockchartR", "quantmod")

Option 1 - In-house ($115k) Hire: Financial Networks Analyst Salary: $115k ($90k + $25k Benefits, Taxes and Insurance) System Requirements: R → IDE RStudio Packages: igraph, stockchartR & quantmod Cost: $0 (open source software)

Option 2 - Outsourcing ($150,600 w/setup cost: $25k) Data-driven Financial Network Analytics Corp (DFNA) Costs: Setup $25k | $150k Annual Fee | $150/quarterly report

That you approve our investment framework and authorize an initial investment of $500mm.

• Please let us know if you have questions. ???
• We've already moved the money into our account.
• and placed trades.
• we are down at the moment but don't worry.