Most distribution free nonparametric methods depend on the ranks or orderings of the individual observations. This dissertation develops methods for the situation when there is only partial information about the ranks available. A random-linear-extension exact test and an empirical version of the random-linear-extension test are proposed as a new way to compare groups of data with partial orders. The basic computation procedure is to generate all possible permutations constrained by the known partial order using a randomization method similar in nature to multiple imputation. This random-linear-extension test can be simply implemented using a Gibbs Sampler to generate a random sample of complete orderings. Given a complete ordering, standard nonparametric methods, such as the Wilcoxon rank-sum test, can be applied, and the corresponding test statistics and rejection regions can be calculated. As a direct result of our new method, a single p-value is replaced by a distribution of p-values. This is related to some recent work on Fuzzy P-values, which was introduced by Geyer and Meeden in Statistical Science in 2005. A special case is to compare two groups when only two objects can be compared at a time. Three matching schemes, random matching, ordered matching and reverse matching are introduced and compared between each other. The results described in this dissertation provide some surprising insights into the statistical information in partial orderings. / Statistics

Polar codes, introduced by Arikan in 2009, have attracted considerable interest as an asymptotically capacity-achieving code with sufficient performance advantages to merit inclusion in the 5G standard. Polar codes are constructed directly from an explicit model of the communication channel, so their performance is dependent on a detailed understanding of the transmission environment. We partially remove a basic assumption in coding theory that channels are identical and independent by extending polar codes to several types of channels with memory, including periodic Markov processes and Information Regular processes. In addition, we consider modifications to the polar code construction so that the inclusion of a shared secret in the frozen set naturally produces encryption via one-time pad. We describe one such modification in terms of the achievable frozen sets which are compatible with the polar code automorphism group. We then provide a partial characterization of these frozen sets using an explicit construction for the Linear Extension Diameter of channel entropies. / Doctor of Philosophy / Efficient, reliable communication has become an essential component of modern society. Error-correcting codes allow for the use of redundant symbols to fix errors in transmission. While it has long been known that communication channels have an inherent capacity describing the optimal redundancy required for reliable transmission, explicit constructions which achieve this capacity have proved elusive. Our focus is the recently discovered family of polar codes, which are known to be asymptotically capacity-achieving. Polar codes also perform well enough in practice to merit inclusion in the 5G wireless standard shortly after their creation. The polarization process uses an explicit model of the channel and a recursive construction to concentrate errors in a few symbols (called the frozen set), which are then simply ignored. This reliance on an explicit channel model is problematic due to a long-standing assumption in coding theory that the probability of error in each symbol is identical and independent. We extend existing results to explore persistent sources of interference modelling environments such as nearby power lines or prolonged outages. While polar codes behave quite well in these new settings, some forms of memory can only be overcome using very long codewords. We next explore an application relating to secure communication, where messages must be recovered by a legitimate receiver but not by an eavesdropper. Polar codes behave quite well in this environment as well, as we can separately compute which symbols can be recovered by each party and use only those with the desired properties. We extend a recent result which proposes the use of a shared secret in the code construction to further complicate recovery by an eavesdropper. We consider several modifications to the construction of polar codes which allow the shared secret to be used for encryption in addition to the existing information theoretic use. We discover that this task is closely related to the unsolved problem of determining which symbols are in the frozen set for a particular channel. We conclude with partial results to this problem, including two choices of frozen set which are, in some sense, maximally separated.

The development of coral reefs is largely restricted to areas within the tropics where favourable conditions for both coral and reef growth prevail. There is, however, a continuum from these typical, accretive reefs in the tropics to marginal, non-accretive, coral-dominated reef communities which occur at higher latitudes. High-latitude reefs function similarly in many regards to their tropical counterparts and are regulated by similar processes to a varying degree. In this study, the major biological and physico-chemical processes were assessed which directly or indirectly prevent the continued persistence of reefal frameworks and thus hinder reef accretion on high-latitude reefs in the iSimangaliso Wetland Park. These reefs have a high diversity of hard and soft corals with significant reef coverage, yet little evidence of any biogenic accretion has been observed. The scleractinian coral, Acropora austera, is one of the few corals which may be responsible for reef framework production. It exhibits a gregarious growth pattern, forming large, monospecific stands with an interlocking framework characteristic of the early stages of reef accretion. The framebuilding potential of A. austera and the continued persistence of such frameworks were thus determined by in situ monitoring of coral growth, mortality, bioerosion and several physico-chemical parameters. Growth rate and mortality of A. austera branches were measured at three sites of differing stand size and apparent age. This was achieved by repeated image analysis and by staining branches with the vital stain, Alizarin Red S. Both measures of growth yielded a similar linear extension rate of 24.5 mm/yr (n = 467), comparable to related species at similar latitudes. Mean branch mortality was as high as 50%, with clear differences manifested between each A. austera stand. Branch extension rates and branch mortality were inversely related between sites. Small, young stands exhibited significantly faster coral growth rates, lower mortality and a net increase in overall branch length over the study period, whilst the opposite was true of larger, more developed stands. In addition, bioerosion was determined at each site to assess its potential for carbonate removal and its destabilizing effect on reef frameworks. Bioerosion intensity was recorded as “percentage area damage” within cross-sections and “frequency of occurrence” of bioeroding organisms in coral rubble fragments (n = 120). The level of bioerosion was found to be substantial (up to 11.5% loss in weight of coral fragments over the 12-month study period) and was found to decrease significantly with a reduction in size of each A. austera stand. Aragonite saturation state is considered a major factor that limits the geographical range of coral reefs globally. Although previously thought to be limiting in Maputaland, mean ΩArag values of 4.40±0.29 were measured on the reefs in summer and 4.33±0.21 in winter and thus would not have limited reef development. Past studies have noted the turbulence on South African east coast reefs and its adverse effect on reef development. This was corroborated in this study with the measurement of considerable sediment re-suspension (0.17 g cm⁻² day⁻¹) and regular damage to both living coral and the reef framework caused by large swells. These results lead to the theory that Acropora austera stands senesce with increasing size and age. Although large coral frameworks are found on the Maputaland reefs, they do not persist in the long term. High rates of sediment re-suspension prevent infilling of the interstitial spaces and eventual cementation, while high levels of bioerosion lead to framework instability over time. Rough seas further hamper accretion by physical removal of both living coral and the coral-derived framework, thus removing recent growth. This process is suspected to cause an imbalance in the carbonate budget of these marginal reefs, ultimately favoring carbonate removal over carbonate deposition. / Thesis (M.Sc.)-University of KwaZulu-Natal, Westville, 2011.

Let <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D%20=%20(%5Cmathbb%7BP%7D,%20%5Cpreceq)" data-classname="equation_inline" data-title="" /> be a finite poset (partially ordered set) with cardinality <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?n" data-classname="equation_inline" data-title="" />. A linear extension of <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D" data-classname="equation_inline" data-title="" /> is an order-preserving bijection <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Csigma" data-classname="equation_inline" data-title="" />: <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D%20%5Crightarrow%20%5Bn%5D" data-classname="equation_inline" data-title="" />, that is, if <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?x%20%5Cpreceq%20y" data-classname="equation_inline" data-title="" /> in <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D" data-classname="equation_inline" data-title="" /> then <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Csigma(x)%20%5Cle%20%5Csigma(y)" data-classname="equation_inline" data-title="" />. We define the poset probability <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?P(%5Calpha%20%5Cpreceq%20%5Cbeta)" data-classname="equation_inline" data-title="" /> as the proportion of linear extensions where <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Csigma(%5Calpha)%20%5Cle%20%5Csigma(%5Cbeta)" data-classname="equation_inline" data-title="" />. We are primarily interested in <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?P(%5Calpha%20%5Cpreceq%20%5Cbeta)" data-classname="equation_inline" data-title="" /> for incomparable elements <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Calpha%20%5Cparallel%20%5Cbeta" data-classname="equation" data-title="" />. The probability has significance in areas such as information theory. Let <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?e(%5Cmathbb%7BP%7D)" data-classname="equation_inline" data-title="" /> denote the total number of linear extensions of <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D" data-classname="equation_inline" data-title="" />. We prove that the poset probability can be evaluated as <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?P(%5Calpha%20%5Cpreceq%20%5Cbeta)%20=%20%5Cfrac%7B%20%5Csum_%7BT%20%5Cin%20B(%5Calpha,%5Cbeta)%7D%20e(T)%20e(%5Cmathbb%7BP%7D%20%5Csetminus%20(T%20%5Ccup%20%5C%7B%5Calpha%5C%7D))%7D%7Be(%5Cmathbb%7BP%7D)%7D" data-classname="equation" data-title="" /> where <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?B(%5Calpha,%5Cbeta)" data-classname="equation_inline" data-title="" /> is the set of order ideals of <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D" data-classname="equation_inline" data-title="" /> without <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Calpha" data-classname="equation" data-title="" /> or <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cbeta" data-classname="equation" data-title="" />, where we can add <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Calpha" data-classname="equation_inline" data-title="" /> to get a new order ideal of <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Cmathbb%7BP%7D" data-classname="equation_inline" data-title="" />. The practicality of the preceding formula is explored and we show that <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?T%20%5Cin%20B(%5Calpha,%5Cbeta)%20%5CLeftrightarrow%20%5Cleft%5C%7B%20x%20%7C%20x%20%5Cprec%20%5Calpha%20%5Cright%5C%7D%20%5Csubseteq%20T%20%5Ctext%7B%20and%20%7D%20T%20%5Ctext%7B%20order%20ideal%20of%20%7D%0A%5Cleft%5C%7B%20x%20%7C%20%5Calpha%20%5Cnot%20%5Cpreceq%20x,%5C%20%5Cbeta%20%5Cnot%20%5Cpreceq%20x%7D" data-classname="equation" /> The formula is particularly useful for certain classes of posets such as partition posets which are examined in further detail. We apply the formula to prove that, for all partition posets of shape <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?%5Bn,n%5D" data-classname="equation_inline" data-title="" />, the probability obeys <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?P((2,a)%20%5Cpreceq%20(1,a+1))%20=%20%5Cfrac%7B%20C_a%20C_%7Bn-a%7D%7D%20%7BC_n%7D" data-classname="equation" data-title="" /> where <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?C_n" data-classname="equation_inline" data-title="" /> is the nth Catalan number and <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?a%20%3C%20n" data-classname="equation_inline" data-title="" />. We also explore how Monte Carlo methods can be used to estimate <img src="http://www.diva-portal.org/cgi-bin/mimetex.cgi?P(%5Calpha%20%5Cpreceq%20%5Cbeta)" data-classname="equation_inline" data-title="" />.