# Dealing With Multicollinearity: A Brief Overview and Introduction to Tolerant Methods

This semester I’m taking a Multivariate statistics course taught by Professor Scott Steinschneider in the BEE department at Cornell. I’ve been really enjoying the course thus far and thought I would share some of what we’ve covered in the class with a blog post. The material below on multicollinearity is from Dr. Steinschneider’s class, presented in my own words.

### What is Multicollinearity?

Multicollinearity is the condition where two or more predictor variables in a statistical model are linearly related (Dormann et. al. 2013). The existence of multicollinearity in your data set can result in an increase of the variance of regression coefficients leading to unstable estimation of parameter values. This in turn can lead to erroneous identification of relevant predictors within a regression and detracts from a model’s ability to extrapolate beyond the range of the sample it was constructed with. In this post, I’ll explain how multicollinearity causes problems for linear regression by Ordinary Least Squares (OLS), introduce three metrics for detecting multicollinearity and detail two “Tolerant Methods” for dealing with multicollinearity within a data set.

### How does multicollinearity cause problems in OLS regression?

To illustrate the problems caused by multicollinearity, let’s start with a linear regression:

$y=x\beta +\epsilon$

Where:

$y=x\beta +\epsilon$

$x = a \hspace{.1 cm} vector \hspace{.1 cm} of \hspace{.1 cm} predictor \hspace{.1 cm} variables$

$\beta = a \hspace{.1 cm} vector \hspace{.1 cm} of \hspace{.1 cm} coefficients$

$\epsilon = a \hspace{.1 cm} vector \hspace{.1 cm} of \hspace{.1 cm} residuals$

The Gauss-Markov theorem states that the Best Linear Unbiased Estimator (BLUE) for each  coefficient can be found using OLS:

$\hat{\beta}_{OLS} = (x^Tx)^{-1}x^Ty$

This  estimate will have a variance defined as:

$var(\hat{\beta}_{OLS}) =\sigma^2 (x^Tx)^{-1}$

Where:

$\sigma^2 = the \hspace{.1 cm} variance\hspace{.1 cm} of \hspace{.1 cm} the\hspace{.1 cm} residuals$

If you dive into the matrix algebra, you will find that the term (xTx) is equal to a matrix with ones on the diagonals and the pairwise Pearson’s correlation coefficients (ρ) on the off-diagonals:

$(x^Tx) =\begin{bmatrix} 1 & \rho \\ \rho & 1 \end{bmatrix}$

As the correlation values increase, the values within (xTx)-1 also increase. Even with a low residual variance, multicollinearity can cause large increases in estimator variance. Here are a few examples of the effect of multicollinearity using a hypothetical regression with two predictors:

$\rho = .3 \rightarrow (x^Tx)^{-1} =\begin{bmatrix} 1 & \rho \\ \rho & 1 \end{bmatrix}^{-1} = \begin{bmatrix} 1.09 & -0.33 \\ -0.33 & 1.09 \end{bmatrix}$

$\rho = .9 \rightarrow (x^Tx)^{-1} =\begin{bmatrix} 1 & \rho \\ \rho & 1 \end{bmatrix}^{-1} = \begin{bmatrix} 5.26 & -4.73 \\ -5.26 & -4.73 \end{bmatrix}$

$\rho = .999 \rightarrow (x^Tx)^{-1} =\begin{bmatrix} 1 & \rho \\ \rho & 1 \end{bmatrix}^{-1} = \begin{bmatrix} 500.25 & -499.75 \\ -499.75 & 500.25\end{bmatrix}$

So why should you care about the variance of your coefficient estimators? The answer depends on what the purpose of your model is. If your only goal is to obtain an accurate measure of the predictand, the presence of multicollinearity in your predictors might not be such a problem. If, however, you are trying to identify the key predictors that effect the predictand, multicollinearity is a big problem.

OLS estimators with large variances are highly unstable, meaning that if you construct estimators from different data samples you will potentially get wildly different estimates of your coefficient values (Dormann et al. 2013). Large estimator variance also undermines the trustworthiness of hypothesis testing of the significance of coefficients. Recall that the t value used in hypothesis testing for an OLS regression coefficient is a function of the sample standard deviation (the square root of the variance) of the  OLS estimator.

$t_{n-2} =\frac{\hat{\beta_j}-0}{s_{\beta_j}}$

An estimator with an inflated standard deviation, $s_{\beta_j}$, will thus yield a lower t value, which could lead to the false rejection of a significant predictor (ie. a type II error). See Ohlemüller et al. (2008) for some examples where hypothesis testing results are undermined by multicollinearity.

### Detecting Multicollinearity within a data set

Now we know how multicollinearity causes problems in our regression, but how can we tell if there is multicollinearity within a data set? There are several commonly used metrics for which basic guidelines have been developed to determine whether multicollinearity is present.

The most basic metric is the pairwise Pearson Correlation Coefficient between predictors, r. Recall from your intro statistics course that the Pearson Correlation Coefficient is a measure of the linear relationship between two variables, defined as:

$r_{x_1,x_2}=\frac{cov(x_1,x_2)}{\sigma_x\sigma_y}$

A common rule of thumb is that multicollinearity may be a problem in a data set if any pairwise |r| > 0.7 (Dormann et al. 2013).

Another common metric is known as the Variance Inflation Factor (VIF). This measure is calculated by regressing each predictor on all others being used in the regression.

$VIF(\beta_j) = \frac{1}{1-R^2_j}$

Where Rj2 is the R2 value generated by regressing predictor xj on all other predictors. Multicollinearity is thought to be a problem if VIF > 10 for any given predictor (Dormann et al. 2012).

A third metric for detecting multicollinearity in a data set is the Condition Number (CN) of the predictor matrix defined as the square root of the ratio of the largest and smallest eigenvalues in the predictor matrix:

$CN=\sqrt{\frac{\lambda_{max}}{\lambda_{min}}}$

CN> 15 indicates the possible presence of multicollinearity, while a CN > 30 indicates serious multicollinearity problems (Dormann et al. 2013).

### Dealing with Multicollinearity using Tolerant Methods

In a statistical sense, there is no way to “fix” multicollinearity. However, methods have been developed to mitigate its effects. Perhaps the most effective way to remedy multicollinearity is to make a priori judgements about the relationship between predictors and remove or consolidate predictors that have known correlations. This is not always possible however, especially when the true functional forms of relationships are not known (which is often why regression is done in the first place). In this section I will explain two “Tolerant Methods” for dealing with multicollinearity.

The purpose of Tolerant Methods is to reduce the sensitivity of regression parameters to multicollinearity. This is accomplished through penalized regression. Since multicollinearity can result in large and opposite signed  estimator values for correlated predictors, a penalty function is imposed to keep the value of predictors below a pre-specified value.

$\sum_{j=1}^{p}|\beta|^l \leq c$

Where c is the predetermined value representing model complexity, p is the number of predictors and l is either 1 or 2 depending on the type of tolerant method employed (more on this below).

#### Ridge Regression

Ridge regression uses the L2 norm, or Euclidean distance, to constrain model coefficients (ie. l = 2 in the equation above). The coefficients created using ridge regression are defined as:

$\hat{\beta}_{r} = (x^Tx+\lambda I)^{-1}x^Ty$

Ridge regression adds a constant, λ, to the term xTx to construct the estimator. It should be noted that both x and y should be standardized before this estimator is constructed. The Ridge regression coefficient is the result of a constrained version of the ordinary least squares optimization problem. The objective is to minimize the sum of square errors for the regression while meeting the complexity constraint.

$\hat{\beta_r} \begin{cases} argmin(\beta) \hspace{.1cm}\sum_{i=1}^{N} \epsilon_i^2 \\ \sum_{j=1}^{p}|\beta_j|^2 \leq c \end{cases}$

To solve the constrained optimization, Lagrange multipliers can be employed. Let z equal the Residual Sum of Squares (RSS) to be minimized:

$argmin(\beta) \hspace{.3cm} z= (y-x\beta)^T(y-x\beta)+\lambda(\sum_{i=1}^{N}|\beta_j|^2-c)$

This can be rewritten in terms of the L2 norm of β:

$z = (y-x\beta)^T(y-x\beta)+\lambda||\beta||^2_2$

Taking the derivative with respect to β and solving:

$0 = \frac{\partial z}{\partial \beta} = -2x^T(y-x\beta)+2\lambda\beta$

$x^Ty = x^Tx\beta+\lambda\beta=(x^Tx+\lambda I)\beta$

$\hat{\beta}_{r} = (x^Tx+\lambda I)^{-1}x^Ty$

Remember that the Gauss-Markov theorem states that the OLS estimate for regression coefficients is the BLUE, so by using ridge regression, we are sacrificing some benefits of OLS estimators in order to constrain estimator variance. Estimators constructed using ridge regression are in fact biased, this can be proven by calculating the expected value of ridge regression coefficients.

$E[\hat{\beta_r}]=(I+\lambda(x^Tx)^{-1})\beta \neq \beta$

For a scenario with two predictors, the tradeoff between reduced model complexity and increase bias in the estimators can be visualized graphically by plotting the estimators of the two beta values against each other. The vector of beta values estimated by regression are represented as points on this plot  $(\hat{\beta}=[\beta_1, \beta_2])$.  In Figure 1,$\beta_{OLS}$ is plotted in the upper right quadrant and represents estimator that produces the smallest RSS possible for the model. The ellipses centered around  are representations of the increasing RSS resulting from the combination of β1 and β2  values, each RSS is a function of a different lambda value added to the regression.  The circle centered around the origin represents the chosen level of model complexity that is constraining the ridge regression. The ridge estimator is the point where this circle intersects a RSS ellipse. Notice that as the value of c increases, the error introduced into the estimators decreases and vice versa.

Figure 1: Geometric Interpretation of a ridge regression estimator. The blue dot indicates the OLS estimate of Beta, ellipses centered around the OLS estimates represent RSS contours for each Beta 1, Beta 2 combination (denoted on here as z from the optimization equation above). The model complexity is constrained by distance c from the origin. The ridge regression estimator of Beta is shown as the red dot, where the RSS contour meets the circle defined by c.

The c value displayed in Figure 1 is only presented to explain the theoretical underpinnings of ridge regression. In practice, c is never specified, rather, a value for λ is chosen a priori to model construction. Lambda is usually chosen through a process known as k-fold cross validation, which is accomplished through the following steps:

1. Partition data set into K separate sets of equal size
2. For each k = 1 …k, fit model with excluding the kth set.
3. Predict for the kth set
4. Calculate the cross validation error (CVerror)for kth set: $CV^{\lambda_0}_k = E[\sum(y-\hat{y})^2]$
5. Repeat for different values of , choose a that minimizes   $CV^{\lambda_0} = \frac{1}{k}CV^{\lambda_0}_k$

#### Lasso Regression

Another Tolerant Method for dealing with multicollinearity known as Least Absolute Shrinkage and Selection Operator (LASSO) regression, solves the same constrained optimization problem as ridge regression, but uses the L1 norm rather than the L2 norm as a measure of complexity.

$\hat{\beta}_{Lasso} \begin{cases} argmin(\beta) \hspace{.1cm}\sum_{i=1}^{N} \epsilon_i^2 \\ \sum_{j=1}^{p}|\beta_j|^1 \leq c \end{cases}$

LASSO regression can be visualized similarly to ridge regression, but since c is defined by the sum of absolute values of beta, rather than sum of squares, the area it constrains is diamond shaped rather than circular.  Figure 2 shows the selection of the beta estimator from LASSO regression. As you can see, the use of the L1 norm means LASSO regression selects one of the predictors and drops the other (weights it as zero). This has been argued to provide a more interpretable estimators (Tibshirani 1996).

Figure 2: Geometric interpretation of Lasso Regression Estimator. The blue dot indicates the OLS estimate of Beta, ellipses centered around the OLS estimate represents RSS contours for each Beta 1, Beta 2 combination (denoted as z from the optimization equation). The mode complexity is constrained by the L1 norm representing model complexity. The Lasso estimator of Beta is shown as the red dot, the location where the RSS contour intersects the diamond defined by c.

### Final thoughts

If you’re creating a model with multiple predictors, it’s important to be cognizant of potential for multicollinearity within your data set. Tolerant methods are only one of many possible remedies for multicollinearity (other notable techniques include data clustering and Principle Component Analysis) but it’s important to remember that no known technique can truly “solve” the problem of multicollinearity. The method chosen to deal with multicollinearity should be chosen on a case to case basis and multiple methods should be employed if possible to help identify the underlying structure within the predictor data set (Dormann et. al. 2013)

### Citations

Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J. R. G., Gruber, B., Lafourcade, B., Leitão, P. J., Münkemüller, T., McClean, C., Osborne, P. E., Reineking, B., Schröder, B., Skidmore, A. K., Zurell, D. and Lautenbach, S. 2013, “Collinearity: a review of methods to deal with it and a simulation study evaluating their performance.” Ecography, 36: 27–46. doi:10.1111/j.1600-0587.2012.07348.x

Ohlemüller, R. et al. 2008. “The coincidence of climatic and species rarity: high risk to small-range species from climate change.” Biology Letters. 4: 568 – 572.

Tibshirani, Robert 1996. “Regression shrinkage and selection via the lasso.” Journal of the Royal Statistical Society. Series B (Methodological): 267-288.

# Synthetic streamflow generation

A recent research focus of our group has been the development and use of synthetic streamflow generators.  There are many tools one might use to generate synthetic streamflows and it may not be obvious which is right for a specific application or what the inherent limitations of each method are.  More fundamentally, it may not be obvious why it is desirable to generate synthetic streamflows in the first place.  This will be the first in a series of blog posts on the synthetic streamflow generators in which I hope to sketch out the various categories of generation methods and their appropriate use as I see it.  In this first post we’ll focus on the motivation and history behind the development of synthetic streamflow generators and broadly categorize them.

### Why should we use synthetic hydrology?

The most obvious reason to use synthetic hydrology is if there is little or no data for your system (see Lamontagne, 2015).  Another obvious reason is if you are trying to evaluate the effect of hydrologic non-stationarity on your system (Herman et al. 2015; Borgomeo et al. 2015).  In that case you could use synthetic methods to generate flows reflecting a shift in hydrologic regime.  But are there other reasons to use synthetic hydrology?

In water resources systems analysis it is common practice to evaluate the efficacy of management or planning strategies by simulating system performance over the historical record, or over some critical period.  In this approach, new strategies are evaluated by asking the question:  How well would we have done with your new strategy?

This may be an appealing approach, especially if some event was particularly traumatic to your system. But is this a robust way of evaluating alternative strategies?  It’s important to remember that any hydrologic record, no matter how long, is only a single realization of a stochastic process.  Importantly, drought and flood events emerge as the result of specific sequences of events, unlikely to be repeated.  Furthermore, there is a 50% chance that the worst flood or drought in an N year record will be exceeded in the next N years.  Is it well advised to tailor our strategies to past circumstances that will likely never be repeated and will as likely as not be exceeded?  As Lettenmaier et al. [1987] reminds us “Little is certain about the future except that it will be unlike the past.”

Even under stationarity and even with long hydrologic records, the use of synthetic streamflow can improve the efficacy of planning and management strategies by exposing them to larger and more diverse flood and drought than those in the record (Loucks et al. 1981; Vogel and Stedinger, 1988; Loucks et al. 2005).  Figure 7.12 from Loucks et al. 2005 shows a typical experimental set-up using synthetic hydrology with a simulation model.  Often our group will wrap an optimization model like Borg around this set up, where the system design/operating policy (bottom of the figure) are the decision variables, and the system performance (right of the figure) are the objective(s).

(Loucks et al. 2005)

### What are the types of generators?

Many synthetic streamflow generation techniques have been proposed since the early 1960s.  It can be difficult for a researcher or practitioner to know which method is best suited to the problem at hand.  Thus, we’ll start with a very broad characterization of what is out there, then proceed to some history.

Broadly speaking there are two approaches to generating synthetic hydrology: indirect and direct.  The indirect approach generates streamflow by synthetically generating the forcings to a hydrologic model.  For instance one might generate precipitation and temperature series and input them to a hydrologic model of a basin (e.g. Steinschneider et al. 2014).  In contrast, direct methods use statistical techniques to generate streamflow timeseries directly.

The direct approach is generally easier to apply and more parsimonious because it does not require the selection, calibration, and validation of a separate hydrologic model (Najafi et al. 2011).  On the other hand, the indirect approach may be desirable.  Climate projections from GCMs often include temperature or precipitation changes, but may not describe hydrologic shifts at a resolution or precision that is useful.  In other cases, profound regime shifts may be difficult to represent with statistical models and may require process-driven models, thus necessitating the indirect approach.

Julie’s earlier series focused on indirect approaches, so we’ll focus on the direct approach.  Regardless of the approach many of the methods are same.  In general generator methods can be divided into two categories: parametric and non-parametricParametric methods rely on a hypothesized statistical model of streamflow whose parameters are selected to achieve a desired result (Stedinger and Taylor, 1982a).  In contrast non-parametric methods do not make strong structural assumptions about the processes generating the streamflow, but rather rely on re-sampling from the hydrologic record in some way (Lall, 1995).  Some methods combine parametric and non-parametric techniques, which we’ll refer to as semi-parametric (Herman et al. 2015).

Both parametric and non-parametric methods have advantages and disadvantages.  Parametric methods are often parsimonious, and often have analytical forms that allow easy parameter manipulation to reflect non-stationarity.  However, there can be concern that the underlying statistical models may not reflect the hydrologic reality well (Sharma et al. 1997).  Furthermore, in multi-dimensional, multi-scale problems the proliferation of parameters can make parametric models intractable (Grygier and Stedinger, 1988).  Extensive work has been done to confront both challenges, but they may lead a researcher to adopt a non-parametric method instead.

Because many non-parametric methods ‘re-sample’ flows from a record, realism is not generally a concern, and most re-sampling schemes are computationally straight forward (relatively speaking).  On the other hand, manipulating synthetic flows to reflect non-stationarity may not be as straightforward as a simple parameter change, though methods have been suggested (Herman et al. 2015Borgomeo et al. 2015).  More fundamentally, because non-parametric methods rely so heavily on the data, they require sufficiently long records to ensure there is enough hydrologic variability to sample.  Short records can be a concern for parametric methods as well, though parametric uncertainty can be explicitly considered in those methods (Stedinger and Taylor, 1982b).  Of course, parametric methods also have structural uncertainty that non-parametric models largely avoid by not assuming an explicit statistical model.

In the coming posts we’ll dig into the nuances of the different methods in greater detail.

### A historical perspective

The first use of synthetic flow generation seems to have been by Hazen [1914].  That work attempted to quantify the reliability of a water supply by aggregating the streamflow records of local streams into a 300-year ‘synthetic record.’  Of course the problem with this is that the cross-correlation between concurrent flows rendered the effective record length much less than the nominal 300 years.

Next Barnes [1954] generated 1,000 years of streamflow for a basin in Australia by drawing random flows from a normal distribution with mean and variance equal to the sample estimates from the observed record.  That work was extended by researchers from the Harvard Water Program to account for autocorrelation of monthly flows (Maass et al., 1962; Thomas and Fiering, 1962).  Later work also considered the use of non-normal distributions (Fiering, 1967), and the generation of correlated concurrent flows at multiple sites (Beard, 1965; Matalas, 1967).

Those early methods relied on first-order autoregressive models that regressed flows in the current period on the flows of the previous period (see Loucks et al.’s Figure 7.13  below).  Box and Jenkins [1970] extended those methods to autoregressive models of arbitrary order, moving average models of arbitrary order, and autoregressive-moving average models of arbitrary order.  Those models were the focus of extensive research over the course of the 1970s and 1980s and underpin many of the parametric generators that are widely used in hydrology today (see Salas et al. 1980; Grygier and Stedinger, 1990; Salas, 1993; Loucks et al. 2005).

(Loucks et al. 2005)

By the mid-1990s, non-parametric methods began to gain popularity (Lall, 1995).  While much of this work has its roots in earlier work from the 1970s and 1980s (Yakowitz, 1973, 1979, 1985; Schuster and Yakowitz, 1979; Yakowitz and Karlsson, 1987; Karlson and Yakowitz, 1987), improvements in computing and the availability of large data sets meant that by the mid-1990s non-parametric methods were feasible (Lall and Sharma, 1996).  Early examples of non-parametric methods include block bootstrapping (Vogel and Shallcross, 1996), k-nearest neighbor (Lall and Sharma, 1996), and kernel density methods (Sharma et al. 1997).  Since that time extensive research has made improvement to these methods, often by incorporating parametric elements.  For instance, Srinivas and Srinivasan (2001, 2005, and 2006) develop a hybrid autoregressive-block bootstrapping method designed to improve the bias in lagged correlation and to generate flows other than the historical, for multiple sites and multiple seasons.  K-nearest neighbor methods have also been the focus of extensive research (Rajagopalan and Lall, 1999; Harrold et al. 2003; Yates et al. 2003; Sharif and Burn, 2007; Mehortra and Sharma, 2006; Prairie et al. 2006; Lee et al. 2010, Salas and Lee, 2010, Nowak et al., 2010), including recent work by our group  (Giuliani et al. 2014).

Emerging work focuses on stochastic streamflow generation using copulas [Lee and Salas, 2011; Fan et al. 2016], entropy theory bootstrapping [Srivastav and Simonovic, 2014], and wavelets [Kwon et al. 2007; Erkyihun et al., 2016], among other methods.

In the following posts I’ll address different challenges in stochastic generation [e.g. long-term persistence, parametric uncertainty, multi-site generation, seasonality, etc.] and the relative strengths and shortcomings of the various methods for addressing them.

### Works Cited

Barnes, F. B., Storage required for a city water supply, J. Inst. Eng. Australia 26(9), 198-203, 1954.

Beard, L. R., Use of interrelated records to simulate streamflow, J. Hydrol. Div., ASCE 91(HY5), 13-22, 1965.

Borgomeo, E., Farmer, C. L., and Hall, J. W. (2015). “Numerical rivers: A synthetic streamflow generator for water resources vulnerability assessments.” Water Resour. Res., 51(7), 5382–5405.

Y.R. Fan, W.W. Huang, G.H. Huang, Y.P. Li, K. Huang, Z. Li, Hydrologic risk analysis in the Yangtze River basin through coupling Gaussian mixtures into copulas, Advances in Water Resources, Volume 88, February 2016, Pages 170-185.

Fiering, M.B, Streamflow Synthesis, Harvard University Press, Cambridge, Mass., 1967.

Giuliani, M., J. D. Herman, A. Castelletti, and P. Reed (2014), Many-objective reservoir policy identification and refinement to reduce policy inertia and myopia in water management, Water Resour. Res., 50, 3355–3377, doi:10.1002/2013WR014700.

Grygier, J.C., and J.R. Stedinger, Condensed Disaggregation Procedures and Conservation Corrections for Stochastic Hydrology, Water Resour. Res. 24(10), 1574-1584, 1988.

Grygier, J.C., and J.R. Stedinger, SPIGOT Technical Description, Version 2.6, 1990.

Harrold, T. I., Sharma, A., and Sheather, S. J. (2003). “A nonparametric model for stochastic generation of daily rainfall amounts.” Water Resour. Res., 39(12), 1343.

Hazen, A., Storage to be provided in impounding reservoirs for municipal water systems, Trans. Am. Soc. Civ. Eng. 77, 1539, 1914.

Herman, J.D., H.B. Zeff, J.R. Lamontagne, P.M. Reed, and G. Characklis (2016), Synthetic Drought Scenario Generation to Support Bottom-Up Water Supply Vulnerability Assessments, Journal of Water Resources Planning & Management, doi: 10.1061/(ASCE)WR.1943-5452.0000701.

Karlsson, M., and S. Yakowitz, Nearest-Neighbor methods for nonparametric rainfall-runoff forecasting, Water Resour. Res., 23, 1300-1308, 1987.

Kwon, H.-H., U. Lall, and A. F. Khalil (2007), Stochastic simulation model for nonstationary time series using an autoregressive wavelet decomposition: Applications to rainfall and temperature, Water Resour. Res., 43, W05407, doi:10.1029/2006WR005258.

Lall, U., Recent advances in nonparametric function estimation: Hydraulic applications, U.S. Natl. Rep. Int. Union Geod. Geophys. 1991- 1994, Rev. Geophys., 33, 1093, 1995.

Lall, U., and A. Sharma (1996), A nearest neighbor bootstrap for resampling hydrologic time series, Water Resour. Res. 32(3), pp. 679-693.

Lamontagne, J.R. 2015,Representation of Uncertainty and Corridor DP for Hydropower 272 Optimization, PhD edn, Cornell University, Ithaca, NY.

Lee, T., J. D. Salas, and J. Prairie (2010), An enhanced nonparametric streamflow disaggregation model with genetic algorithm, Water Resour. Res., 46, W08545, doi:10.1029/2009WR007761.

Lee, T., and J. Salas (2011), Copula-based stochastic simulation of hydrological data applied to Nile River flows, Hydrol. Res., 42(4), 318–330.

Lettenmaier, D. P., K. M. Latham, R. N. Palmer, J. R. Lund and S. J. Burges, Strategies for coping with drought Part II: Planning techniques for planning and reliability assessment, EPRI P-5201, Final Report Project 2194-1, June 1987.

Loucks, D.P., Stedinger, J.R. & Haith, D.A. 1981, Water Resources Systems Planning and Analysis, 1st edn, Prentice-Hall, Englewood Cliffs, N.J.

Loucks, D.P. et al. 2005, Water Resources Systems Planning and Management: An Introduction to Methods, Models and Applications, UNESCO, Delft, The Netherlands.

Maass, A., M. M. Hufschmidt, R. Dorfman, H. A. Thomas, Jr., S. A. Marglin and G. M. Fair,

Design of Water Resource Systems, Harvard University Press, Cambridge, Mass., 1962.

Matalas, N. C., Mathematical assessment of synthetic hydrology, Water Resour. Res. 3(4), 937-945, 1967.

Najafi, M. R., Moradkhani, H., and Jung, I. W. (2011). “Assessing the uncertainties of hydrologic model selection in climate change impact studies.” Hydrol. Process., 25(18), 2814–2826.

Nowak, K., J. Prairie, B. Rajagopalan, and U. Lall (2010), A nonparametric stochastic approach for multisite disaggregation of annual to daily

streamﬂow, Water Resour. Res., 46, W08529, doi:10.1029/2009WR008530.

Nowak, K., J. Prairie, B. Rajagopalan, and U. Lall (2010), A nonparametric stochastic approach for multisite disaggregation of annual to daily

streamﬂow, Water Resour. Res., 46, W08529, doi:10.1029/2009WR008530.

Nowak, K., J. Prairie, B. Rajagopalan, and U. Lall (2010), A nonparametric stochastic approach for multisite disaggregation of annual to daily

streamﬂow, Water Resour. Res., 46, W08529, doi:10.1029/2009WR008530.

Nowak, K., J. Prairie, B. Rajagopalan, and U. Lall (2010), A nonparametric stochastic approach for multisite disaggregation of annual to daily streamflow, Water Resour. Res., 46, W08529, doi:10.1029/2009WR008530.

Prairie, J. R., Rajagopalan, B., Fulp, T. J., and Zagona, E. A. (2006). “Modified K-NN model for stochastic streamflow simulation.” J. Hydrol. Eng., 11(4), 371–378.

Rajagopalan, B., and Lall, U. (1999). “A k-nearest-neighbor simulator for daily precipitation and other weather variables.” Water Resour. Res., 35(10), 3089–3101.

Salas, J. D., J. W. Deller, V. Yevjevich and W. L. Lane, Applied Modeling of Hydrologic Time Series, Water Resources Publications, Littleton, Colo., 1980.

Salas, J.D., 1993, Analysis and Modeling of Hydrologic Time Series, Chapter 19 (72 p.) in The McGraw Hill Handbook of Hydrology, D.R. Maidment, Editor.

Salas, J.D., T. Lee. (2010). Nonparametric Simulation of Single-Site Seasonal Streamflow, J. Hydrol. Eng., 15(4), 284-296.

Schuster, E., and S. Yakowitz, Contributions to the theory of nonparametric regression, with application to system identification, Ann. Stat., 7, 139-149, 1979.

Sharif, M., and Burn, D. H. (2007). “Improved K-nearest neighbor weather generating model.” J. Hydrol. Eng., 12(1), 42–51.

Sharma, A., Tarboton, D. G., and Lall, U., 1997. “Streamflow simulation: A nonparametric approach.” Water Resour. Res., 33(2), 291–308.

Srinivas, V. V., and Srinivasan, K. (2001). “A hybrid stochastic model for multiseason streamflow simulation.” Water Resour. Res., 37(10), 2537–2549.

Srinivas, V. V., and Srinivasan, K. (2005). “Hybrid moving block bootstrap for stochastic simulation of multi-site multi-season streamflows.” J. Hydrol., 302(1–4), 307–330.

Srinivas, V. V., and Srinivasan, K. (2006). “Hybrid matched-block bootstrap for stochastic simulation of multiseason streamflows.” J. Hydrol., 329(1–2), 1–15.

Roshan K. Srivastav, Slobodan P. Simonovic, An analytical procedure for multi-site, multi-season streamflow generation using maximum entropy bootstrapping, Environmental Modelling & Software, Volume 59, September 2014a, Pages 59-75.

Stedinger, J. R. and M. R. Taylor, Sythetic streamflow generation, Part 1. Model verification and validation, Water Resour. Res. 18(4), 909-918, 1982a.

Stedinger, J. R. and M. R. Taylor, Sythetic streamflow generation, Part 2. Parameter uncertainty,Water Resour. Res. 18(4), 919-924, 1982b.

Steinschneider, S., Wi, S., and Brown, C. (2014). “The integrated effects of climate and hydrologic uncertainty on future flood risk assessments.” Hydrol. Process., 29(12), 2823–2839.

Thomas, H. A. and M. B. Fiering, Mathematical synthesis of streamflow sequences for the analysis of river basins by simulation, in Design of Water Resource Systems, by A. Maass, M. Hufschmidt, R. Dorfman, H. A. Thomas, Jr., S. A. Marglin and G. M. Fair, Harvard University Press, Cambridge, Mass., 1962.

Vogel, R.M., and J.R. Stedinger, The value of stochastic streamflow models in over-year reservoir design applications, Water Resour. Res. 24(9), 1483-90, 1988.

Vogel, R. M., and A. L. Shallcross (1996), The moving block bootstrap versus parametric time series models, Water Resour. Res., 32(6), 1875–1882.

Yakowitz, S., A stochastic model for daily river flows in an arid region, Water Resour. Res., 9, 1271-1285, 1973.

Yakowitz, S., Nonparametric estimation of markov transition functions, Ann. Stat., 7, 671-679, 1979.

Yakowitz, S. J., Nonparametric density estimation, prediction, and regression for markov sequences J. Am. Stat. Assoc., 80, 215-221, 1985.

Yakowitz, S., and M. Karlsson, Nearest-neighbor methods with application to rainfall/runoff prediction, in Stochastic  Hydrology, edited by J. B. Macneil and G. J. Humphries, pp. 149-160, D. Reidel, Norwell, Mass., 1987.

Yates, D., Gangopadhyay, S., Rajagopalan, B., and Strzepek, K. (2003). “A technique for generating regional climate scenarios using a nearest-neighbor algorithm.” Water Resour. Res., 39(7), 1199.

# New Federal Flood Frequency Analysis Guidelines

This post is a bit off topic for this blog, but I think it should be interesting to our readers.  The current procedure used by Federal agencies (FEMA, Bureau of Reclamation, USGS, Army Corps, etc.) to assess flood risk at gauged sites is described in Bulletin 17B.  That procedure is used to estimate flood risk for things like FEMA flood maps, to set flood insurance rates, and to design riparian structures like levees.  Given how important the procedure is, many people are surprised to learn that it has not been updated since 1982, despite improvements in statistical techniques and computational resources, and the additional data that is now available.  This post wont speculate as to why change has taken so long, but I will briefly describe the old procedures and the proposed updates.

Bulletin 17B Procedure:

Dr. Veronica Webster has written an excellent brief history on the evolution of  national flood frequency standards in the US dating back to the 1960s.  For this post, we focus on the procedures adopted in 1982.  In short, Bulletin 17B recommends fitting the log-Pearson Type III distribution to the annual peak flow series at a gauged location, using the log-space method of moments.  In other words, one takes the logarithms of the flood series and then uses the mean, variance, and skew coefficient of the transformed series to fit a Pearson Type III distribution.  The Pearson Type III distribution is a shifted Gamma distribution.  When the population skew coefficient is positive the distribution is bounded on the low end, and when the population skew is negative the distribution is bounded on the high end.  When the population skew is zero, the Pearson Type III distribution becomes a normal distribution.  For those wanting more information, Veronica and Dr. Jery Stedinger have written a nice three-part series on the log-Pearson Type III, starting with a paper on the distribution characteristics.

Unfortunately data is not always well behaved and the true distribution of floods is certainly not log-Person Type III, so the Bulletin takes measures to make the procedure more robust.  One such measure is the use of a regional skew coefficient to supplement the sample skew coefficient computed from the transformed flood record.  The idea here is that computing the skew coefficient from short samples is difficult because the skew coefficient can be noisy, so one instead uses a weighted average of the sample skew and a regional skew.  The relative weights are proportional to the mean squared error of each skew, with the more accurate skew given more weight.  The Bulletin recommends obtaining a regional skew from either an average of nearby and hydrologically similar records, a model of skew based on basin characteristics (like drainage area), or the use of the National Skew Map (see below), based on a 1974 study.

National Skew Map provided in Bulletin 17B [IAWCD, 1982, Plate I]

A second concern is that small observations in a flood record might exert unjustifiable influence in the analysis and distort the estimates of the likelihood of the large floods of interest.  Often such small observations are caused by different hydrologic processes than those that cause the largest floods.  In my own experience, I’ve encountered basins wherein the maximum annual discharge is zero, or nearly zero in some years.  The Bulletin recommends censoring such observations, meaning that one removes them from the computation of the sample moments, then applies a probability adjustment to account for their presence.  The Bulletin provides an objective outlier detection test to identify potential nuisance observations, but ultimately leaves censoring decisions to the subjective judgement of the analyst.  The proposed objective test is the Grubbs-Beck test, which is a 10% significance test of whether the smallest observation in a normal sample is unusually small.

The Hydrologic Frequency Analysis Work Group (HFAWG) is charged with updating the Bulletin.  Their recommendations can be found on-line, as well as a testing report which compares the new methods to the old Bulletin.  Bulletin 17C is currently being written.  Like Bulletin 17B, the new recommendations also fit the log-Pearson Type III distribution to the annual maximum series from a gaged site, but a new fitting technique is used: the expected moments algorithm (EMA).  This method is related to the method of moments estimators previously used, but allows for the incorporation of historical/paleo flood information, censored observations, and regional skew information in a unified, systematic methodology.

High water mark of 1530 flood of the Tiber River at Rome

Historical information might include non-systematic records of large floods: “The town hall has been there since 1760 and has only been flooded once,”  thus providing a threshold flow which has only been crossed once in 256 years.  EMA can include that sort of valuable community experience about flooding in a statistically rigorous way! For the case that no historical information is available, no observations are censored, and no regional skew is used, the EMA moment estimators are exactly the same as the Bulletin 17B method of moments estimators.  See Dr. Tim Cohn’s website for EMA software.

The EMA methodology also has correct quantile confidence intervals (confidence intervals for the 100-year flood, etc.), which are more accurate than the ones used in Bulletin 17B.

Another update to the Bulletin involves the identification of outlier observations, which are now termed Potentially Influential Low Flows (PILFs).  A concern with the old detection test was that it rarely identified multiple outliers, even when several very small observations are present.  In fact, the Grubbs-Beck test, as used in the Bulletin, is only designed to test the smallest observation and not the second, third, or kth smallest observations.  Instead, the Bulletin adopts a generalization of the Grubbs-Beck test designed to test for multiple PILFs (see this proceedings, full paper to come).  The new test identifies PILFs more often and will identify more PILFs than the previous test, but we’ve found that use of a reasonable outlier test actually results in improved quantile efficiency when fitting log-Pearson Type III data with EMA  (again, full paper in review).

The final major revision is the retirement of the skew map (see above), and the adoption of language recommending more modern techniques, like Bayesian Generalized Least Squares (BGLS).  In fact, Dr. Andrea Veilleux, along with her colleagues at the USGS have been busy using iterations of BGLS to generate models of regional skew across the country, including California, Iowa, the Southeast, the Northwest, MissouriVermont, Alaska, Arizona, and California rainfall floods.  My masters work was in this area, and I’d be happy to write further on Bayesian regression techniques for hydrologic data, if folks are interested!

If folks are interested in software that implements the new techniques, the USGS has put out a package with good documentation.

That’s it for now!

Jon

# C++ Training: Exercise 1

(Updated 1/30/12)

This example requires some data you can download here: inputdata.  Open the file in Excel and “Save As”… csv.

Write a C++ program that reads data from the csv file and then calculates the mean, the standard deviation, the 10th percentile, and the 90th percentile, of each column of data.  Output these statistics to one or more new files.  You may create one file for each statistic, i.e.

means-of-data.csv

col1, col2, col3,
mean-of-col-1, mean-of-col-2, mean-of-col-3

Or put each statistic in its own row in a single file.

For this exercise, you may need the following functions, libraries, or features.  Please look them up on cplusplus.com and make liberal use of the example code there!

math: pow, operators such as +, -, and *

input/output of data: please use ifstream and ofstream.  You’ll want to use the << and >> operator, and the function: getline

manipulating text streams: the c++ library string, and stringstream, may be helpful.  You may need to look up how to handle csv files, and how to separate the commas from the data

data: sort

control structures: if, else, while, for

One piece of advice: if you want to convert a C++ style string to an integer or vice versa, you may find the following functions handy:


string intToString(int input_int)

{

string s;

stringstream out;

out << input_int;

s = out.str();

out.clear();

return s;

}

int stringToInt(string input_string)

{

return atoi(input_string.c_str());

}

double stringToDouble(string input_string)

{

return strtod(input_string.c_str(), NULL);

}



Unit 1: Getting the Program Working

1. Write a standard deviation function that can take the standard deviation of an arbitrary number of values. Verify that the function works either with a calculator, Matlab, or Excel. Is the precision the same? What could cause differences between your answer and the calculators?
2. Learn how to use the input/output streams.  You’ll want to set up one stream to read the file, and another one to write output files. To test this, you can simply “copy” the file you read in directly into another file. Are there ways to do this without using too much system memory?
3. The easiest statistics to calculate are mean and standard deviation, since they can be done without storing much information.  The 10th and 90th percentile are more difficult, because you must sort the values first.  Don’t try to do these until you have the other two tasks completed.
4. Once you have your results, verify them using Matlab or Excel.  Are your statistics correct?

Unit 2: Intermediate Steps